1. Home
  2. Samples List
  3. View

US Election Stock Final V4

2016 US election and its influence on the stock market
Contents
Introduction ..................................................................................................................................... 4
Background ................................................................................................................................. 4
Purpose of the research ............................................................................................................... 7
Hypothesis................................................................................................................................... 7
Structure of the Thesis ................................................................................................................ 8
Literature Review.......................................................................................................................... 10
The Presidential Puzzle ............................................................................................................. 12
Political Business Cycles .......................................................................................................... 13
Presidential Election Cycle ....................................................................................................... 15
Partisan Theory ......................................................................................................................... 21
Efficient Market Hypothesis (EMH) ........................................................................................ 23
Statistical Definitions ................................................................................................................ 25
Time Series Data and Time Series Analysis ......................................................................... 25
P-value .................................................................................................................................. 26
Hypothesis Testing................................................................................................................ 26
Multiple Linear Regression Analysis.................................................................................... 27
The Impact of Politics on Investors .......................................................................................... 28
Methodology ................................................................................................................................. 31
Models for Normal Returns ...................................................................................................... 35
Statistical Models .................................................................................................................. 35
Economic Models ................................................................................................................. 35
The Market Model ................................................................................................................ 36
Abnormal Returns ..................................................................................................................... 36
Research Design........................................................................................................................ 37
The Action Plan ........................................................................................................................ 40
Data selection and collection .................................................................................................... 41
Variables ............................................................................................................................... 42
Dummy Variable ................................................................................................................... 51
Analysis and Discussion ............................................................................................................... 52
ADF Testing.............................................................................................................................. 54
Correlogram Analysis ............................................................................................................... 57
Cointegration Test ..................................................................................................................... 58
Conclusion and Future Research .................................................................................................. 62
References ..................................................................................................................................... 65
Appendix ....................................................................................................................................... 72
Introduction
Background
The year 2016 was one of the most scandalous years in recent history, loaded with attention and
debates vis-à-vis American Presidential campaign. The biggest reason behind all the frenzy was
the fact that both the candidates running for the US President Office 2016 were considered to be
quite controversial. The political climate of the US has been gaining widespread attention but it is
not restricted to the country alone. During that period and even now, the attention given to the
political environment of countries such as Denmark, England and Sweden has been increasing
significantly. Their right-wing parties are known to be enjoying boosted support which has startled
many political figures across the world. Politics and all the associated decision making has a
significant on the functioning of the society of a country. This is the reason why politics and
elections are considered to be a popular subject that constantly attracts attention.
Political uncertainty has an innate ability to influence a country’s stock market considerably.
Elections are those events through which the investors can obtain important information for
investment purposes. Previous literature on the subject of the impact of the national elections on
the stock market of a country revealed that the surrounding political uncertainty has an impact on
the beliefs of the investors (Bialkowski, Gottschalk and Wisniewski, 2008). Whenever there is an
election to be held in a country, the volatility and stock prices are observed to rise drastically. The
research primarily focused on the developed countries. Research on the US Presidential elections
in 2016 and its impact on the stock markets will be helpful for investors who are looking forward
to distribute investments or already hold a collection of assets in the US or a country that is prone
to elections shock in near future.
The equity markets and Presidential election have always been notable topics for research papers
and the main reason for this favoritism is the fact that the policies set up by the President influence
the ability of the businesses and the overall economy to grow. The historic events are proof that
generally businesses prefer to endorse some political agenda that favors their objectives and goals
(Caro, 2002). Sometimes it is seen that the businesses are ready to invest huge sum of money to
back a political agenda or particular candidate because whoever will win the elections will also
impact the business environment directly. Therefore, relationship between the politics and the
stock market cannot be neglected. However, the connection between the economic conditions and
election results cannot be identified clearly at all times because the basic economic conditions such
as hardship have a considerable effect on the voting results through public opinion (Fiorina, 1991).
There are two ways through which the election results can affect the business environment of the
country and both rely on influencing the economy of the country. The first one attempts to do so
by making changes in government expenditure/fiscal changes while the second one changes the
regulatory environment of the market after new administration starts operation. The American
political structure is directed by two big parties, namely, the Republicans and the Democrats. By
and large, Republicans are known to be inclined towards laisser-faire capitalism and favor
deregulation and low taxes. On the other hand, the Democrats are known to make use of a
Keynesian approach that leverages government to act as a facilitator of socio-economic progress.
However, the impact is not distributed evenly across the stock market which can be both favorable
and detrimental to a particular company/sector.
In the United States, Presidential elections are the major factor that changes the political landscape
of the country. Before an election, traders involve their expectations in relation to the political
changes in the stock prices. After the election results, adjustments are done according to the results
and the subsequent decision-making. The increasing prospects of the candidate’s triumph is
supposed to reflect in the stock prices. But, the speculations surrounding the election outcomes are
not always unambiguous. Thus, tight elections result in an increase in the volatility of the futures
markets because of the uncertainty with reference to election outcomes and associated
consequences (Jones, 2008). An information asymmetry exists between the political parties and
market until the Election Day. Political soapboxing is a major factor due to which market
participants get clouded as they cannot distinguish between facts and political posturing (He et al.,
2009).
It is generally believed that the Democratic Presidential administration is good for the stock
market. In view of several researchers, for nearly twenty years, Democratic Presidential
administration is known to guarantee hearty market returns (Allvine and O’Neill, 1980). In
addition, the discoveries of this research recommend that the political cycle in the case of stock
returns is to a great degree huge and has remained stable over a great part of the previous century.
The anticipated difference between the Democratic and Republican administrations regarding the
excess returns is around 9 percent per year for large stocks and more than 16 percent per year for
small stocks. In addition, previous studies have revealed that difference in the returns throughout
the presidential cycles is large, however there is no proof of differences in the market risk across
the Republican and Democratic administrations. Santa-Clara and Valkanov (2003) have termed
this as a Presidential Puzzle.
Goodell and Vähämaa (2013) showed that the election process develops market uncertainty as
financial specialists create assumptions in regards to would-be leaders and future economic
policies. As soon as the probability of a candidate winning the election increases, the volatility
starts decreasing (Goodell & Bodey, 2012). This means that the market reacts negatively to the
decreasing P/E ratios.
Purpose of the research
The study aims to determine the impact of the 2016 US election on the stock market. The political
agenda behind the presidential, parliamentary or legislative elections is generally observed to make
changes in the policies that have the ability to change the economic environment. Also, the election
cycle in the countries establishes patterns for government and investment behavior. And the
elections can cause political and social uncertainty significantly. The paper aims to contribute to
the existing literature in relation to the effect of the presidential elections in the US on the stock
market performance of the US industry sectors and stock corporations. The determination and
analysis of the connection between political system and stock market performance are quite
difficult. It involves an analysis of the political alignment of the parties. Some parties can be more
inclined towards the right-wing side of the political balance while some parties can be biased
towards the left-wing side. The study focuses on the US Presidential elections because the US is
one of the biggest economy across the globe and the biggest stock exchange, New York Stock
Exchange (NYSE) is established in the United States. The 2016 US election saw many investments
analysts forecasting increased ebb and flow in the stock prices. All these rationales make it
interesting to research on the underlying subject in the US.
Hypothesis
The study has developed two hypotheses to determine the extent to which the stock market
changed after the 2016 US presidential election. The hypotheses have been formulated on the basis
of the assumption that a stock market development is an outcome of the perspective of the public
regarding the future economic state of the country. This implies that if people foresee a positive
future economy, the stock market shows growth. On the contrary, if the people are negatively
doubtful about the future, the stock market decreases.
H1: There is a correlation between the US presidential election and the stock market performance
which is revealed through the changes in the stock prices after the election.
H2: After the 2016 US election, the stock market exhibited a positive growth and industry such as
finance, raw materials, technology and manufactured goods demonstrated high levels of growth.
Structure of the Thesis
Chapter 1 (Introduction and Background): This will be the first chapter that will introduce the
topic of the research to the audience. The chapter will set out the background to the situation and
also the aims and objectives of the research.
Chapter 2 (Literature Review): This chapter will focus on the existing literature and theories in
the area and form the backbone of the research. Particular focus will be had on theories that are
associated with explaining stock market shocks and those that are specifically linked to elections
such as the Presidential Election Cycle Theory. The aim of this chapter is to test and add to the
existing theories.
Chapter 3 (Methodology): This chapter will reveal the methodology that would be employed for
testing the aims and objectives of the research which is the determination of the 2016 US election
on the stock market prices.
Chapter 4 (Data): This chapter will draw down the data that will be used for the testing purposes.
Primarily, the data will include the stock market movement over the years ideally over several
decades including periods where there have been other elections to look for a trend and to allow
for the adjustment mentioned in the methodology.
Chapter 5 (Analysis): The central chapter of the research which will take the discussion back to
the original aims and objectives of looking at the impact of the 2016 election specifically. By using
the data gathered and the event study it is anticipated that the research will then be able to look at
the various elements of the impact that has been had on the stock market both immediately after
the election and how it is settling down a year later.
Chapter 6 (Conclusion): Finally overall conclusions will be drawn in a short concise manner that
could effectively be read stand-alone by anyone who wants to hear the findings without going
through the methodology and full data analysis.
Chapter 7 (Summary, Limitations and Future Recommendations): As a final chapter the
dissertation will identify any limitations in the research that would potentially jeopardize the
reliability and thus making future recommendations for other research in this area. For example, it
is anticipated that this will include the recognition that it may be necessary to review the stock
market in another year as the impact of the election is likely to go beyond the current time horizon.
Literature Review
The US presidential elections are considered to be one of the major events that occur every four
years and influences the economies across the world. Election results are observed to affect the
stock market by means of changes in the government policies like tax and spending changes. In
addition, certain sector and industries might profit or loss from governmental policies that are
sector-specific. There are many researchers that have come up with the notion that a small-cap
stock is more likely to get affected by the events like national elections that have an economic
impact. According to a study, the excess returns from such stocks were higher under left-wing
Democratic presidents in the United States (Booth and Booth, 2003). These observations were
substantiated by the findings that confirmed the impact on the small-cap firms. Moreover, in the
cases of large-cap firms, the returns were more or less identical under both Republican and
Democratic administrations (Hensel and Ziemba, 1995).
In the opinion of Hensel and Ziemba, the reason behind this could be fact that a more liberal
government focuses on promoting smaller businesses. The study conducted by Booth and Booth
(2003) upheld a theory of political business cycle based on the patterns found in relation to small-
cap stock returns and large-cap stock returns with a different presidential cycle in the United States.
The defense sector witnessed an improved stock price performance when the chances of Ronald
Reagan winning the 1980 presidential election increased (Roberts, 1990). However, the overall
stock market was not largely affected. For the duration of the campaign period for 1992
presidential election, 15 out of 74 sectors were influenced by the changes in the winning prospects
of the presidential candidates (Herron et al., 1999). Economic sectors along with forecasting equity
prices while keeping election results under consideration, also forecast commodity prices,
currencies and interest rates (Snowberg, Wolfers and Zitzewitz, 2007). Political partisanship
affects the macro economic variables like unemployment, growth and inflation rates for the US
and also for Europe (Caporale and Grier, 2000). It is important to note that the relationship between
the political and economic sphere works both ways as the stock market also facilitate economic
policy reforms by means of measures such as lobbyism (Gerber, Huber and Washington, 2009).
An article “Election Years Roil Markets with Waves of Unease,” in New York Times in May
questioned that whether a plunge in merger and acquisition activity, GDP, market volatility, and a
decrease in spending can be explained by Donald Trump and Hillary Clinton (Sorkin, 2016). It
was noticed that big companies like Exxon Mobil, Delta Air Lines, McDonald’s and Verizon
dramatically cut their financial spending. What’s more, the firms decreased their investment
expenditures by 4.8 percent on an average in comparison to the non-election years (Julio and Yook,
2012). Similarly, in the first quarter of the year 2016, merger volume was down 38 percent in
comparison to the last year (Dealogic, 2016).
The previous research revealed that throughout the Presidential election years and particularly
those with large uncertainty about the future of the country. The stock market gets paralyzed
because the investments and large deals are postponed. 2016 US presidential election was also
such an election that was full of uncertainties and controversies. Historically, the market is known
to trend downwards in the last year of the Presidents’ term which IPOs and investments under
psychological damper. According to a report, Standard & Poor’s Index has decreased by 2.8
percent on an average during the election years wherein the President doesn’t seek re-election
(Stephen Suttmeier, 2016). Out of the President term, the last year is the only one in which negative
returns are generated on an average. In comparison, for the years President stands for reelection,
the average returns for S&P is 12.6 percent on an average and the average for the period 1928-
2014 is around 7.5 percent. A mix of these discoveries and the way that the 2016 elections have
Republican candidate, one not at all like any individual who has run earlier (between Trump's
political inexperience and his racist election campaign) raises a critical question: Will we observe
an impact on the stock market in line with the previous trend or the impact will be a distinct one
due to 2016 US elections involving atypical candidate and abnormal activities)?
The Presidential Puzzle
Since the 1980s, many types of research have contested that a stable and large difference in returns
can be noticed across Republican and Democratic presidential administrations (Santa-Clara and
Valkanov, 2003). The main evidence that existed to favor a political cycle of stock returns was the
average stock returns over different sample periods across Republican and Democratic
presidencies. The research highlighted that the stock market performed better across Democratic
administrations in comparison to the Republican administrations. The findings were puzzling
because the Republicans are considered to be more business-friendly than Democrats. Republicans
have always been vouching for less regulation and low taxes which is beneficial for the stock
market and the underlying shareholders. For quantifying the difference, the Dow Jones Industrial
Average can be analyzed over the decades. Since 1900, DJIA has increased by 3% in an average
year under Republican Presidents and for a Democratic President, the same percentage is around
7%. If a sum of money is growing at 3% a year, then it would take around 23 years to double in
value. On the other hand, money at 7% will double itself in around 10 years. So, there is a big
difference.
Several researchers have cited risk-taking approach, higher inflation and increased spending of the
Democratic presidents to be the reason for the sizable difference. The article “Presidential Puzzle:
Political Cycles and the Stock Market” by Pedro Santa-Clara & Rossen Valkanov examined the
correlation between the political alignment of the incumbent presidential candidate and stock
market performance of the US. The data revealed that the stock market returns were higher when
the incumbent presidential party was Democrat. The writers argued that the difference could be
due to the higher risk premium throughout the Democratic incumbencies. Higher risk premiums
were required as the traders might be worried about the left wing policies. However, this theory
was rejected because the market was observed to be more volatile for Republican administrations.
Santa Clause Clara and Valkanov ended their paper with a basic inquiry in regards to the likelihood
of the possibility that the stock market performance has an impact on which presidential candidate
will be picked. They didn’t arrive at a well-founded conclusion about the motivation to why
Democrats appeared to get higher returns. Rather, they addressed the issue as "Presidential
Puzzle".
Political Business Cycles
This is a hypothesis that expresses that there is a connection between the stock market and the
political occupant party; the officeholder party will or won't be re-chosen relying upon how well
the stock market and the economy is performing. This theory is opposite to what this research is
about as this research aims to evaluate the impact of the US elections on the stock market.
However, it will be important to discuss political business cycles as it will help to gain
understanding on the subject. In the opinion of William D. Nordhaus, unemployment rates and
inflation affects investments. If the unemployment rate increases, there will be a dip in investments
just because of the reason that the households won’t have money to squander. He also believes
that people are more inclined towards steady prices and are opposed to high inflation.
People tend to choose that party which acknowledges their needs. It is generally seen that the
public lacks deep knowledge and thus they depend on the past behavior of the parties for voting
purposes. If the existing government has not been able to improve the economic growth of the
country, then voters will more likely vote in favor of the opposing party. This is the reason why,
most parties attempt to improve the economy and allow the inflation to rise and decrease the
unemployment rate, when the election is to be held in near future. Nordhaus wrote that the
Democrats were more concerned about unemployment while Republicans were more focused on
inflation. According to him, the mainstream business cycle works in the following manner. After
the election results get out, the incumbent party raises the unemployment rate for fighting the
inflation. When the elections are near, the incumbent party lowers the unemployment rate. The
visual representation of the political business cycle according to Nordhaus can be seen from the
following graph:
According to Douglas Hibbs, the change in the unemployment rate and inflation also depends on
the orientation of the government: right wing or left wing. A relatively higher unemployment rate
and low inflation are favored by the higher income class. On the contrary, the low-income class
will prefer low unemployment rate which means higher inflation. Therefore, the high-income class
will support right-wing parties and the low-income class will support left-wing parties.
Presidential Election Cycle
Kitchin (1923) noticed that for the period 1890-1922, a 40 month business cycle was present in
the stock markets of both Great Britain and the United States. Since 40 months is generally the
same length as a President's tenure, he named the pattern to be Presidential Election Cycle. Many
scientists have discovered that the same cycle is still present today. For instance, Stovall (1992)
saw that the Presidential Election Cycle in the US had a particularly huge impact on the stock
markets for the period 1868-1945. The business cycle was characterized by lower returns for the
initial two years of the Presidential administration and significant yields over the last two years of
the President’s term.
In an academic article, "Mapping the Presidential Election Cycle in US Stock Markets," the
researchers Wong and McAleer (2009) put forward that one explanation behind this underlying
decrease in stock costs is that the new a Presidential administration may roll out unpopular or
disagreeable policies for adjusting the economy. In any case, by midterm, stock costs will ascend
because of a now more grounded and improved economy. In a push to evaluate these perceptions,
they observed the week by week information from the Standard and Poor's 500 Composite Price
Index for January 1, 1965- December 31st, 2003 in a bid to analyze empirically. When the index
was graphed, it trend was found to be similar as mentioned before i.e. a four-year Presidential
cycle with prices decreasing for the first two years, attaining a trough for year two and at last,
increasing for the last two years. The researchers also made use of the Exponential Generalized
Autoregressive Conditional Heteroscedasticity model or EGARCH wherein dummy variables
were assigned to each year of the Presidential administration. There was a time variable assigned
to the trend for the period 1965-2000 and a dummy variable was assigned to the existing political
party. After this, Augmented Dickey-Fulley transformation was applied to the stock returns of
S&P. The data was then plotted through a periodogram for revealing the peaks more clearly. The
plotting affirmed that there are many business cycles in the stock prices but the most prominent
were the 4 year Presidential Election Cycle because the biggest surges in the stock prices were
observed to be placed after 200 weeks or 4 years on an average. Thus, they confirmed that there is
a relationship between the stock market and the Presidential election cycle.
The most well-known cycle is the 4 year cycle that is otherwise called the Presidential Election
Cycle or the Kitchin Wave. The hypothesis that tries to clarify the connection between Presidential
elections and stock prices are known to be the Theory of the Presidential Election Cycle. At regular
intervals of four years, American voters choose a President. Because of the President's mind-
boggling impact on both world issues and domestic affairs, the gradually expanding influences of
Presidential races are astounding. Therefore, it must not come as an unexpected info that stock
prices that are known as the leading indicator of the economy, are influenced by the Presidential
elections. It is outstanding that business flourishes in a domain of stable economic policies and
lower taxes, which is correctly what all the policy-makers offer in an election year when not a
single public authority wants to be seen as a supporter of higher tax-and-spend. In this way,
markets have a tendency to flourish in the election years.
According to Kitchen, a 40 month cycle could be witnessed in the US and Great Britain for the
period 1890-1922. The four year presidential cycle was an extremely big element of the stock
market for the period 1868-1945. If a President’s term is observed, it is noticed that in the first two
years low returns in the stock market are obtained while for the next two years high returns are
generated in the market. The Presidential Election Theory was developed by a historian Yale
Hirsch. He was of the opinion that after every 4 years, the Presidential elections leave a huge
impact on the stock market and thus the economy (Hirsch, Y., 2010). He maintained that
throughout the election years, the existing administration aims to boost the economic growth
through different policies and actions like tax cuts and higher spending. Until the period of the
1940s, several industrialized countries allowed the economies to self-adjust but with the
acceptance of Keynesian economics, the governments started to improve the economy with the
help of monetary and fiscal policies. If the theory is broken down, it simply implies that the stock
market is lowest in the following year of the US presidential election. According to the theory, the
stock market will improve only when the next cycle begins for the next US presidential election.
Booth and Booth (2003) examine the U.S. stock market for the period 1926-1996 and they were
the first to thoroughly consider the business conditions while searching for the electoral cycle. To
begin with, through simple average returns on the large and small stocks over years of political
cycles, they affirmed the outcomes of the previous investigations, i.e. electoral cycle exists. The
contrasts between returns in an overabundance of T-Bill rate in the first two and the second two
years of the presidential tenure are observed to be greater for the small stocks (Johnson et al.,
1999).
The investors were making use of the Presidential Election Cycle Theory to time their investments
and to formulate investment strategies and plans around the same. The theory gave out specific
performance predictions for each year of the Presidential term.
President
Year
Change (%)
Hoover
1932
-14.8
FDR
1936
27.9
FDR
1940
-15.1
FDR
1944
13.8
Truman
1948
-0.7
Truman
1952
11.8
Eisenhower
1956
2.6
Eisenhower
1960
-3.0
JFK/Johnson
1964
13.0
Johnson
1968
7.7
Nixon
1972
15.6
Nixon/Ford
1976
19.1
Carter
1980
25.8
Reagan
1984
1.4
Reagan
1988
12.4
Bush I
1922
4.5
Clinton
1996
20.3
Clinton
2000
-10.1
Bush II
2004
9.0
Bush II
2008
-38.5
Obama
2012
13.4
Average (All year 4s)
5.53
S&P Average Presidential Cycle Returns: 1928-2015
Percent of time up (%)
54.5
59.1
81.8
71.4
66.7
Year 1: The Presidential Election Cycle Theory expresses that in the first year of the term of a new
President, he aims to divert the attention of the public from the just concluded presidential
campaign by ensuring that the he really achieves all that he said he'd do amid his campaign. This
is why the stock markets are weakest for the first year of the president term’s first year. Many
anticipated that this same situation would play out after Trump's latest triumph, yet the business
sectors have puzzled analysts by doing precisely the opposite.
Year 2: According to the theory, the second year for all intents and purposes reflects the first year
since the new president is yet doing everything to make sure that all the promises that were made
during the campaign are fulfilled as he is under constant surveillance of the public. His attention
is exclusively on revising current tax laws and by and large, being on the best conduct. Because of
this reason, the stock markets are in a condition of general weakness even now.
Year 3: With elections approaching again, year three witnesses the president make a special effort
in order to improve the economy by all means. The result of which is a surge in the stock markets
which makes it the most profitable year for all the investors in the market.
Year 4: The attention of the president is exclusively on the financial reforms and the employment
rate. Also, with the elections approaching, the investors are the biggest benefactors in the fourth
year as the surge in the market continues to increase and the markets are almost at their highest
performance levels.
So what does the evidence reveal? If the S&P 500 Index for the years 1969, 1977 and 1981
decreased and these were the first years for the Presidential term of Richard Nixon, Jimmy Carter
and Ronald Reagan. In each case of these administrations, the stock market recovered from the
first year jitters and the positive results were seen in the subsequent years. However, the theory
has been rejected by many financial analysts as it has not been able to predict the first year
performance of the market under a new administration. Case in point, when Barack Obama
assumed the office in 2009, the SPX increased by 23.45%. For the year 1993, the first year under
President Bill Clinton, the SPX increased by 10%. Such growth in the SPX was also evident from
the growth in stocks under the first year of President George H.W. Bush.
Since 1944, the stock market rose by 6.2% on an average in the first year of the Presidential term,
as evident from a report by S&P Global Market Intelligence. If the average growth value of the
market is seen for the same time period, the value comes out to be 8.6%. So, it can be seen that in
recent times the stock market has done good in the first year of a new President.
Despite the fact that this hypothesis turned out to be very precise for the early period of time, it
began to prove false for the elections and Presidents towards the finish of the 20
th
century, with
business sectors indicating solid growth following the presidential terms of George Bush and Bill
Clinton. What's more, the pattern appears to have proceeded after Donald Trump's unexpected
triumph in the as of late concluded presidential elections with business sectors witnessing their
best closing figures since 2011.
Partisan Theory
It has been around two decades that there has been consistent and broad academic interest for the
impacts of partisanship on the stock exchanges in the United States. This research and literature
body has firmly been based on partisan cycle model, according to which the economic
performances can be traced back to the strategic actions of the parties. On the basis of the
Downsian perspective of the democracy (Downs 1957), Hibbs (1977) maintains that economic
policies can be related to the ideologies of the parties. Distinctive belief systems infer diverse
financial strategies profiting a few sections of the electorate and detriment of others. In the opinion
of Hobbs, governments formulate macro economic policies mostly on the basis of the subjective
preferences and economic interests of the core class defined political constituencies (Hibbs, 1977).
Therefore, the parties are thought to be ideologically spurred and to adhere to their electoral stages
while holding the office, left-wing parties will try to decrease unemployment rate in the pre-
election phase on the grounds that their voter base advantages more from a lower unemployment
rate than from lower inflation. In addition, every party will always follow their policy goals and
objectives in understanding with their ideologies which direct that inflation will be lower under
right-wing government than left-wing government (Alesina, 1987; Downs, 1957). The study
conducted by Hibbs was based on the data collected from twelve North American and West
European countries. The traditional partisan theory puts forward that the impact of the economic
elements is permanent and if the rational partisan theory is to believe, the impact is temporary.
In a democracy, the result of an election data on the political structure of the ensuing government.
In view of the partisan theory, this ought to take into account the prediction of the future financial
and money related strategies. As left-wing government parties are expected to seek lower
joblessness and the right-wing government parties bring down inflation, financial specialists
anticipate that left-wing will advocate demand-side policies while supply-side policies are
advocated by right-wing parties. This infers that the policies of the right-wing parties are
productive for firms and the stock prices increases as the NPV (Net Present Value) of the expected
dividends also increases. In this way, a change of governance from a left-wing party to a right-
wing party would mean a change in the economic policies invigorating stock prices. During the
term of a left-wing government, a higher inflation would also mean low real return rates for the
investors which make investments in stock less appealing. Subjective and quantitative examination
of the manifestos of the parties additionally reveals that the left-wing governments emphasize on
the demand-side policies to help the lower-salary class, and endeavor to redistribute the incomes
through higher tax assessment for high-income class and companies (Budge et al., 2001).
The investors expect different inflation level for different government administrations which
impacts how much of the volume of trade is done in the stock markets (Leblang and Mukherjee,
2005). According to them, expected higher inflation under left-wing party will amount to low
trading volumes and hence the volatility and mean of the stock prices will also decrease in
comparison to the right-wing parties. They also suggest that this outcome is not only relevant for
the Presidential term but also a certain oriented party is more likely to win the Presidential
elections. The support for the partisan approach and their theory was obtained by them from the
daily and monthly information from the British and US markets. Riley and Luksetich (1980) also
obtained empirical support for high returns under Republican presidents.
However, the empirical confirmation found is not consistent. Gärtner and Welleshoff (1995)
explored the Presidential Election Cycle and noticed that the cycle was available during both right-
wing administrations and left-wing administrations. What's more, yet the profits don't vary by a
huge degree for the both administrations. Prior to this, Huang (1985) had demonstrated that the
Presidential Election Cycle is seen in both the administrations and the returns are more or less
same for the two. However, he also mentioned that when the profits were significantly at variance,
they have really been higher in cases of Democratic administrations.
Efficient Market Hypothesis (EMH)
The efficient market hypothesis is a theory which deals with different forms of types of power on
the market. According to the theory, it would be difficult to beat the market because the stock
prices can completely reveal the important information about the market. In the opinion of Byström
(2014), there are three distinct forms of efficiencies: Strong, Semi-market and Weak. Strong
efficiency means that a certain stock price will reveal all the information in relation to the stock at
once, irrespective of the fact that the information is private or public. Due to this, insider trading
cannot produce lucrative results. Semi-strong efficiency means that all of the public information
is revealed in the stock prices. In such cases, traders can make profits with the help of insider
trading. The weak efficiency means that the stock prices fluctuate at random and there is no
reflection of the historic data.
The idea of the Efficient Market Hypothesis (EMH) advocates that the prices of the financial assets
reveal all the important information. In this manner, stock prices are accurate on average which
means that the financial market is efficient. An immediate outcome is that an active financial
specialist cannot beat the market continuously and the not-so active investors can accomplish
similar profits as an active investor on an average. Generally, market values are constantly true
and the future prices are dependent on the random new information.
The EMH hypothesis was created by Eugen Fama, in the mid-1960. Along with other authors, he
supported a generally overlooked proposal from a French mathematician by the name of Louice
Bachelier. In addition, Eugen Fama broadened and refined the hypothesis with proposing three
types of market efficiency; the Strong form, semi-strong form and weak form (Fama, 1970).
An illustration (Shleifer, 2000) ought to sum up the possibility of the efficient markets. This
illustration is in view of an economic principle called arbitrage. It implies that the simultaneous
buying and selling of a single security in two different stock markets at favorable extraordinary
costs. In this sense, stocks could end up noticeably overrated in respect to its basic value, in case
silly investor purchase these stocks. The stock prices surpass the NPV of the cash flow expected.
At that point reasonable investors or arbitrageurs would sell these stocks as they are aware of the
overpricing and would all the while buying different securities for the adjustment in their profit-
risk ratio. This impact results in a decrease in the stock prices to their basic values. If there is a
quick arbitrage then substitutes are accessible and arbitrageurs compete with one another. These
contentions incorporate that arbitrageurs couldn't acquire unusual benefits. The argument is that
arbitrageurs were not able to gain abnormal profits.
In addition, irrational arbitrageurs that buy overpriced securities and also simultaneously offer
underpriced securities obtain low returns in comparison to the passive investors and therefore lose
money. In the end, the irrational investors lose wealth and leave the market. Therefore, in the long
run, the stock markets tend to be efficient because of the competitive selection and arbitrage
(Shleifer, 2000).
In the cases of different stock markets, the market efficiency is also different depending on
different factors such as size. This theory is also known as Random Walk Theory suggests that
current stock prices have the ability to reveal all the available information on the value of a firm
and it cannot be used by the traders to obtain excess profits. Obtaining profits from predicting the
stock price movements is a very difficult task. The main catalyst that is responsible for the changes
in the price is the existence of new information. An efficient market is that market that wherein
the stock prices make quick adjustments to the new information available without bias. Therefore,
the present costs of securities mirror all accessible data at any given point in time. Subsequently,
there is no motivation to trust that costs are too low or too high.
Security costs change before a trader has time for trading and benefit from another new piece of
information. As indicated by capital markets hypothesis, the anticipated return from securities is
principally a component of the risk. The cost of the security mirrors the present estimation of the
projected future cash-flows, which involves many factors, for example, liquidity, unpredictability
and danger of insolvency. But if the prices are established rationally, price changes are observed
to be unpredictable and random. This is because of the fact that the new information by virtue of
its nature, is also unpredictable. Hence, the stock prices follow the Random Walk.
Statistical Definitions
Time Series Data and Time Series Analysis
Time series data can be considered that type of type which is obtained and analyzed for a certain
period of time (Korner and Wahlgren, 2013). In between the data or observations, there are certain
spaces that is generally equal. For example, collection of data on a monthly basis for the period of
one year. Generally, the time series data is used while examining the closing value of some market
index. When the study of the data is done at same time within a period, it results into data sample
with time series. The data in the time series tracks the changes for a specific period or ratio. In this
study, the change for a certain period is utilized and the data corresponds to the monthly figures.
P-value
In the past for hypothesis testing, the significance level was evaluated before the study was
executed. In present modern times, this level is determined with the help of the observation of the
p-value. It is the probability that a result will be obtained with a value that is larger or equal to the
observed value in the research if the null hypothesis stands true (Korner and Wahlgreen, 2006). In
case the value of the p-value is small, then the null hypothesis is rejected and smaller the value of
the p-value, it is more likely that the alternate hypothesis stands correct.
It is usually seen that the limit in relation to the rejection or acceptance of the null hypothesis is
kept around 5%. However, this value can show variations on the basis of the study that is being
conducted. For example, the study conducted by a researcher for studying the risks related to the
dysfunctional car brake can make use of the low limit p-value to ensure that the brakes are
functional. If the value of the p-value is greater than the 5% limit, there is no importance and thus
in such cases it is unlikely to eliminate a null hypothesis.
Hypothesis Testing
The aim of performing a hypothesis testing is to determine if there are sufficient evidences in the
data sample so that the assumptions can be applied to the total population (Korner and Wahlgren,
2006). The hypothesis testing comes with two assumptions, out of which one assumption will
prove to be true while the other assumption will be untrue and is rejected by the research. The two
facets of the hypothesis testing are known as null-hypothesis and alternate hypothesis. In case, the
null hypothesis gets rejected, it usually indicates that the topic under study is true and that there is
some effect. The result of this testing is dependent on the p-value which has been described above.
Foe our research, the null hypothesis is that there is no impact of the US Presidential Elections on
the performance of the stock market.
Multiple Linear Regression Analysis
The primary motive of performing the regression analysis is related to the investigation of some
relation between an explanatory variable and dependent variable. The most common regression
analysis technique is called the simple linear regression analysis. In this form of linear regression
analysis, there are only 2 variables that are involved. The multiple linear regression analysis takes
the simple linear regression to the next level. In this form of linear regression analysis, instead of
evaluating the relation between one independent and one dependent variable, this analysis involves
two or more number of explanatory variables for the linear regression (Korner and Wahlgren,
2006). Following is an example of multiple linear regression when it is written in an equation form:
[Y= β0 + β1 * x1 + β2 * x2 + β3 * dv1 + β4 * dv2 + …. + βn * xn + β (n + 1) * dvn + Ɛ]
In this case, Y is a dependent variable that is to be studied by the researcher. β0 is the obtained
intercept on the execution of the linear regression analysis. All of the other values of β are
coefficients related to the independent variables (x1+…+xn) and dummy variables (dv1+…+dvn),
wherein n indicates the last sequence number and for different number of variables, n can have
different number. The error term are represented by the Ɛ symbol. It is the summation of all the
values of the residuals. Residual is the estimated prices minus the observed prices for a specific
period.
The Impact of Politics on Investors
In the article “Red and Blue Investing: Values and Finance” by Hong and Kostovetsky (2012), the
US Mutual Fund managers and their stock holdings are observed. Often, these managers donate to
the Presidential campaign of the Democrats and it also impacts their portfolio. The observation
revealed that the managers who actively donated to the Democrat side or the Democratic campaign
were not generally interested in a stock that were socially irresponsible like guns and tobacco.
Such stocks are generally linked with the Republicans. On the other hand, Democratic managers
are more inclined towards stocks of the socially responsible businesses. The observation was same
for the hedge fund managers.
With a specific end goal to assert that politics impact investments, it is essential to not exclusively
just observe the trend in the political party and portfolio allocation but also the adjustments and
amount of investments throughout an election. There is a great deal of financial instability in the
United States economy, particularly political vulnerability because of the economic recession. In
an insightful article of Brandon Julio and Youngsuk Yook (2012), they attempted to evaluate the
impact of uncertainty in relation to the investment expenses by building up two main points. In the
first place, election results are dependent on the corporate decisions since their outcomes will
directly affect industry control, tax assessment, and financial approach. Secondly, in light of the
fact that it is so difficult to evaluate the effect of political uncertainties on investing because of
endogeneity amongst economic growth and uncertainty as a consequence of financial downturn
itself. Elections are basically a test to study the political impacts since they unravel some of the
endogeneity. In a specimen of few huge nations, it was found that investments are at lower values
before the elections. The control over the economic environment and investment opportunities
lead to a 4.8% drop in the investment rates in the year before an election. This observation backs
the hypothesis of political uncertainty which expresses that uncertainty results into firms to
decrease the investment expenses until the election results come out. Aggregately, this
examination demonstrates that politics have an effect on the investment decisions of the firm.
Moreover, the most simple estimation model employed for the interpretation of the movements in
the prices reveals that actual stock prices are equal to the current value of the expected future
dividends with discounted real discount-rate. The model was to a great extent negated by Shiller
(1981) and noticed that prices were excessively unstable and pointed that there is missing data in
this model. As such, the model is excessively oversimplified and Shiller suggests that political
elements might be the key to the unsolved unpredictability. Sweet, Ozimek, and Asher (2016)
clarify that political components can impact securities exchange unpredictability since political
instability can make organizations postpone hiring, firing and finally investments because in order
to gain the system in the stock markets depends on expecting changes, such as Presidential
strategies. On the other side, there is additionally a buyer affect since they might be more
parsimonious amid times of uplifted instability.
However, separating political changes from political uncertainties is very challenging. Many
financial analysts continue to struggle with the complexities of the market behavior and the
incompetence of the present models to completely explain this behavior. Due to this, scholars like
Wisniewski (2009) study the effect of the variables in relation to the politics on the evaluation of
the firms. Wisniewski observed that the behavior of the market was beyond the present value
model. The stocks are more expensive under a Democratic administration. In addition, if there is
strong support backing the President, that period witnesses inflated stock prices. The reason behind
this is that investors are optimistic about the future economic growth. Another factor that makes
investors optimistic is the casting of votes by the voters, as revealed by the increase in prices in
the year of the election. Therefore, the present value model will be accurate only if the political
factors are included.
With these observations and theories under consideration, this research paper seeks to measure the
impact of the 2016 US Presidential election on the stock market of the country. It is challenging
to isolate the political impacts as the 2016 election year contained such unusual candidates running
for the President office. For the Democratic side, the nominee was Hillary Clinton who has more
than 40 years of diplomatic and political experience. On the other hand, the nominee for the
Republicans was Donald Trump who has negligible political experience whatsoever. Therefore,
the volatility in the stock stocks should be unusually high as there would be increased levels of
uncertainties due to Trump’s banter and experience.
Methodology
The methodology aimed towards realizing an event study may at first look appear to be consuming
because the procedures involved in it comprises of many different steps that can take the research
in more than one direction. The complete literature of the event studies can be daunting as the
clarification of each conceivable approach appears to be about incomprehensible. Due to this
reason, the research design of this research paper is primarily focused on the short-term event study
and towards the attributes of the observational examination that has been done the previous chapter
of literature review. Along with all of the different approaches, the undertaking is considerably
more bulky however each progression additionally offers sources of inclination in the measurable
evaluations.
Mackinley (1997) has developed a compact and clear guide that helps in the methodology process
of the event studies. However, he did not explained the different solutions and sources of bias with
a perfect statistical support. Brown and Warner (1980) came up with a different guide for the event
studies wherein they only focused on the different approaches that can be employed while going
through each step and also highlighted the effect of the bias on the different approaches. Their
investigation involved the extraction of different approaches in the research of simulated data with
the help of monthly data. However, in some time, the research was modified to make use of the
daily data too. Also, Bartholdy et al (2016) made use of the event study methodology in relation
to the small stock exchanges wherein thinly trading stocks develop methodological issues. The
research was kept limited for the period 1990-2001 and used daily data from Copenhagen Stock
Exchange. Their findings revealed that the event study methodology employed for small
exchanges can produce valuable results if some adjustments and tunings are made because of thin
trading. Some of the other examples of studies wherein the focus is kept on the different statistical
issues are Corrado (2011), S.V.D. Nageswara Rao and Sreejith (2014), and Binder (1998). Shalit
and Yitzhaki (2002), Saadi et al (2006) and, Lo and MacKinley (1990) kept their focus on a single
bias source in their respective studies. This section will review the methodology engaged to
conduct the event study in order to detect the abnormal returns over the years.
In the literature review, different economists such as Michael McAleer, Wing Keung Wong and
Joseph Kitchin propose that there is a particular four year pattern which can be observed in stock
prices which also reflects the Presidential Election Cycle. Youngsuk Yook and Brandon Julio
explain the importance of the election results on the decisions of the investors. The election results
are known to directly impact the stock market regulations, monetary policies and taxation which
can majorly impact the functioning of the stock markets. In addition, Wisnieswski, Gootschalk
and Bialkowski confirmed that the stock markets of a country react more aggressively in the cases
of closely contested and scandalous elections. The primary problem with these observations is that
the findings are not US specific.
For instance, Wisnieswski, Gootschalk and Bialkowski analyzed election cycles with the help of
data obtained from a sample of around 27 OECD nations and thus the results were not country-
specific. Kitchin conducted a study pertaining to the United States to measure the cycles in stock
prices during the Presidential election in 1923. Despite of the fact that other researchers like Stovall
added extra data to the model proposed by Kitchin, the reality is that the information is more than
60 years old. Another component that remains missing from the research is that it analyzes the
elections on a holistic level and doesn’t keep it centered on the Presidential elections, like in the
research of Julio and Yook.
Allvine and O’Neill (1980) were the first to study that the election cycles were present in the stock
market of the United States. They made use of the monthly returns of the S&P 400 index for the
period 1948-1978. The analysis of the returns presented that there was huge difference between
the returns for the first half and the second half of the election cycle. They found out that the returns
were highest for the third year at 22.1% and for the fourth year were second highest at 9.2%. For
the first and second year, the returns were 0.6 percent and 0.7 percent respectively. Both the
researchers also developed different trading rules that could be employed to exploit the election
cycle to obtain abnormal returns and function in opposition to the market efficiency.
Another methodology was used by Herbst and Slinkman (1984) to determine the impact of the
election dates on the US stock market returns. They looked for the existence of some sinusoidal
patterns of four years and two year periods in relation to the US stock market returns. If the patterns
peaked around the election dates, the effect could be seen as politically induced. Their observations
revealed that the election cycles were existing in the stock market returns for the period 1926-
1977. The two year cycle is at its highest in every nine months on an average and the four year
cycle peaks after them. Herbst and Slinkman concluded that the four year cycle is affected by the
political processes and the short one is because of other reasons.
Bohl and Gottschalk (2004) contributed to the existing literature through making available
international evidence regarding the election cycles in the stock market returns. They created a
dataset that contained information about excess returns on monthly basis from the stock markets
of a set of 15 developed nations. The time series started for the different countries at a different
time period ranging from 1957-1973 with the series ending in 2004. The researchers studied each
country individually and also a panel that included all of the fifteen countries. In addition, they
also analyzed the dataset for the existence of the partisan cycles.
The methodology employed by the authors was same as that of Booth and Booth (2003). Along
with the dividend yields, default and term spreads, they also made use of the expected inflation,
relative interest rate and US stock market returns (lagged value) for the regression purposes. The
theory of the election cycle got rejected for 11 countries and also for the panel. However, some
signs of the election cycle were observed for Netherlands, Canada and Austria. There was a
stronger presence in Sweden. However, the outcomes related to the United States were in
disagreement with the results of the previous studies.
The research by Dopke and Pierdzioch (2006) revolved around the interaction and correlation
between the stock market and the German government. Not following the methodology used in
the previous studies, this research did not take into consideration the impacts of the business cycles
for testing the existence of the election cycle. On the contrary, the capability of the elements of the
business cycle for explaining the political cycles was attested uncertain. However, this model
accounted for the crashes which were the period generating returns that were less than −20%. In
general, the quarterly numbers backed the election cycle theory but very weakly. In fact, the
conclusion made by the researchers was regarding the impact of the stock market variations on the
governments and mainly their popularity.
Wong and McAleer (2009) made use of an entirely different methodology in comparison to the
previous studies. The author utilized the spectral analysis for finding out that strong cycles exist
with a period of 200 weeks on an average which comes out to be four years approximately in the
returns for S&P 500 for the period 1965-2003. They then employed the EGARCH model including
variables that captured the trend evident from the data and some dummy variables. The model was
used for the assessment of the presence of the electoral cycle. The cycle that was determined was
largely in line with the US Presidential elections. The results also revealed that the electoral cycle
is stronger under the Republican administrations and particularly when they are in their second
tenure.
Models for Normal Returns
According to MacKinley, there are two categories of such models, namely economic and statistical
models. The only difference that separates the two models is the use of assumptions from the
economic and statistics models from the investors’ behavior for the modelling of expected returns,
in the case of statistical models. It is necessary to include the statistical assumptions in the
economic model, if it is to be used in practice. The assumptions for statistical models are based on
the fact that the asset returns can be considered as identically and independently distributed, and
multivariate normal. This is sufficient for the Market Model and constant Mean Return Model.
Statistical Models
One of the simplest models is Constant Mean Return Model that makes use of a constant-return
factor and some disturbance term (expected value is 0) for defining the normal return of a stock.
The market model makes use of the return on market for defining the expected returns of a firm.
In addition, the market model is also extendable to the multifactor models that includes other
additional factors other than market return for the explanation of the security returns. However,
the explanatory power do not get much increased by such models in comparison to the simple
market model.
Economic Models
CAPM or Capital Asset Pricing Model is likely the best-known resource return model among the
finance students. The model requires less presentation, however the important part is to relate the
individual returns to the covariance within the market. The correct estimation of the risk or beta
with the help of the CAPM is considered to be problematic. There is an another economic model
that is based on the Arbitrage Pricing Theory (APT) wherein the normal security returns are
determined with the help of the different risk factors. According to multiple researches, market
factor is most effective in the explanation of the expected returns. Thus, APT is not able to provide
significant benefits in relation to market models (Brown and Weinstein, 1985).
The Market Model
In general, market model is considered to be the most valuable model for the estimation of the
normal returns. This model has the highest explanatory power as it can relate individual returns
with the market returns. The expression for the model is:
R
it
= α
i +
β
i
R
mt
+ ε
it
E (ε
it
= 0) var (ε
it
) = σ
2
In this equation, α is the constant factor and β
i
is determined for every individual security
econometrically. There is also some error term included in the security returns. Prior to the
estimation with the help of this model, an estimation window is required to be developed. For
example, MacKinley assigned 250 trading days as the estimation window before event window.
One of the examples of stock market indexes that use this model to estimate returns is S&P 500 in
the US. On completing the collection of the returns on the stocks and the returns on market index,
the model is developed through the OLS methodology for the firm. After this, abnormal returns
for securities is determined in the form of residual terms: the difference between the obtained
returns and the predicted returns through market model. The primary benefit of the market model
over other models is that it removes fragment of returns that is generated due to the variance of
market returns. The greater the value of R
2
, greater are the benefits by the use of market model.
Abnormal Returns
According to MacKinley, abnormal return is the actual return of the securities from an event
window minus normal returns of a firm from an event window. After the estimation of the normal
returns through the market model, abnormal return for the security (i) and an event date (τ) can be
defined by:
AR
iτ
= Riτ – E (R
iτ
| X
τ
)
X
τ
is the information that acts as the condition on which the calculation of the normal returns is
based. The market model defines the Conditional Variance of Abnormal Returns by:
L1 is the estimation window length and if the L increases, the 2
nd
term moves to 0 and Conditional
Variance of Abnormal Returns is approximated through the 1
st
term. It is the squared regression
error for the market model. If a large estimation window is chosen, estimation of the variance of
the abnormal returns needed to test a null hypothesis is rendered unproblematic.
The observation of a series of abnormal returns of a firm does not give information on the interest.
Therefore, there is a need for the aggregation of the abnormal returns for event firm across time.
This results in the production of CAR or cumulative abnormal return of a security. If T1 and T2
are the last days of estimation window and event window, then CAR is estimated from τ1 to τ2
(T1 < τ1 ≤ τ2 ≤ T2). The CAR for a security (i) across an event window can be calculated by:
CAR
i
(τ1, τ2) = Σ AR
iτ
Research Design
Ever since the development of Efficient Market Hypothesis (EMH), it has been researched by
several researchers. In the cases of the rational investors, the share prices reflect the expectation
and information completely. And also, new information is included in the equity prices quickly.
However, empirical researches reveal that the stock prices do not reflect the complete information
most often. The contradiction has led to the development of a new hypothesis in the behavioral
finance called as Uncertain Information Hypothesis or UIH (Brown et. al., 1988). This hypothesis
is based on the assumption that anxiety and uncertainties will increase in the economic markets
due to the happening of some unexpected event. The reason behind is that the investors are not
able to react to the unexpected new information appropriately and therefore initially security prices
are set lesser than the fundamental values. In addition, UIH assumes that stock returns are more
likely to be higher than average returns over period of no uncertainty that are event-induced. In
the cases of uncertainty that are election-induced, a positive outcomes of the cumulative abnormal
returns are expected for the time period following the elections. Therefore, the hypothesis can be
written in the form of:
H1: CAR-
10;0
> 0
H2: CAR
0;+1
> 0
It is generally assumed that after the announcement of the election results, the uncertainty that
surrounds the political event will get resolved. Thus, the stock market needs time to determine the
consequences of the elections after the elections result in an appropriate manner. In case, the
uncertainty after the election results is higher, higher abnormal returns will be noticed post-
election. In short, UIH maintains that the mitigation of the uncertainties would get positive returns
and that wide uncertainty reduction results in great observed returns. If the time period is 10 days
after the date of election, then another hypothesis can be written in the form of:
H3: CAR
+1;+10
> 0
This research paper focuses on the analysis of the 2016 US elections on the stock market of the
country. The result of the election was made known on the 8
th
of November, 2016. This is also
viewed as announcement day. The sample data of this research paper includes information related
to the change in the prices of 3 United States stock price indices. The DJIA or Dow Jones Industrial
Average monitors the prices of thirty broadly traded stocks in the New York Stock Exchange is
seen as the most popular index of the world, however it doesn’t represent the market on the whole.
The other index is Nasdaq Composite that is a capitalization-weighted index for almost 3000
stocks listed on NASDAQ stock exchange. The third index is Standard and Poor’s 500 Composite
stock price index (S&P 500) tracks the performance of US 500 biggest capitalization stocks. Each
of the index includes financials for selected companies (reinsurance, insurance, financial services,
banks), gas and oil (gas and oil producers, oil services and equipment), consumer goods (home
construction, personal goods, household goods), real estate, pharmaceuticals, utilities (water,
power generation, gas, electricity). The data for the NASDAQ stock, S&P 500 and Dow Jones
Industrial Average are available on the OECD database.
This research paper makes use of the Standard Market Model Event study methodology for the
analysis purposes. Prior to conduct the research, it should be pointed out that event studies
investigate the response of the average stock market to some stock market events. The best results
are obtained in relation to the event study are obtained when the accurate event date is identified
or known. The 2016 Presidential election is considered to be the dummy variable. The time period
before the announcement of the election results is given the value “0” and the time period after the
announcement of the election results is given the value “1”. After this, estimation window and
event window are evaluated (Figure 3). The interval which is [T
1+1
, T
2
] is event window and the
length is L
2
= T
2
– T
1-1
. On the other hand, estimation window is restricted to the interval [T
0+1
, T
1
]
and the length for this interval is L
1
= T
1
– T
0-1
. The length for an event window is commonly
dependent on the competency to correctly identify the date of announcement. If a researcher is
able to identify it accurately, than an event window is known to be lesser lengthy and the
procurement of the abnormal return is more effective and precise.
For the purposes of this research, this paper will make use of the monthly stock prices of the three
stock indices for the period surrounding the election result, beginning with the year 2011 and
ending with the year 2017. The months before 2016/11/08 are assigned to be the estimation period
and the following months are assigned to be the event period. The Cumulative Abnormal Returns
or CAR for the event window [τ1; τ2] that surrounds the event day (t=0) wherein [ τ1 ;τ2 ] = ∈ [
−10 ;+10 ] can be shown as:
CAR
i, [τ1, τ2]
= Σ (R
i,t
- α
i
- β
i
R
M, t
)
CAR
i, [τ1, τ2]
is the CAR for share i for the event window [τ1; τ2]. R
i, t
is the actual stock return for
the day t. α and β are the coefficients of regression from the Ordinary Least Squares regression
(OLS) for the estimation period till the t = -10. The event day for the analysis would be the day of
Trump’s victory i.e. 08 November, 2016. After this, the regression analysis will be applied for the
observation of the abnormal returns for the indices.
The Action Plan
The analysis started by downloading and exporting the data into the Excel files. All the data for
each variable was downloaded from the information available on the OECD database, Yahoo
finance and NASDAQ Composite. Each cell in the file contained value for the variables for each
month for the period of 2012/1/1 to 2017/7/1. In addition, there was also some additional data that
came along with the downloaded information. Thus, the undesired data was separated from the
downloaded information such as indicator, subject, measure, frequency and flag codes. All the
data was altered and certain modifications were done in order to make the desired information for
for the research purposes.
After all the data was saved in the Excel files, the software Eviews 9 was used. The Eviews was
to be used for performing the regression analysis. For this, all the data was imported in the Eviews
workfile and all the dependent and independent variables were created. After this, a dummy
variable was also created with value “0” for time period before the election results and value “1”
for time period after the election results. Stock prices were the dependent variables while long-
term interest rates, short-term interest rates and GDP were independent variables. The regression
analysis was carried out on the software with a 95% confidence level. It was seen that the values
of the R-square was very high and the reason behind this was the trend in the downloaded data.
The projected multiple linear regression can be written in the form of:
[Share prices= β0 + β1 * Long-term interest rate + β2 * Short-term interest rate + β3 * GDP + β4
* Dummy Variable (before an election) + β5 * Dummy Variable (after an election)
Data selection and collection
In line with the previous researches, the stock indices and variables to be included in the study are
based on some criteria. First of all, the stock index must have companies that have released their
financial reports for a period of at least 10 consecutive quarters prior to the sample quarter.
Secondly, the companies should have been listed on either of the three indices, DJIA, NASDAQ
Composite or S&P 500 for the least trading days to be 250 before the event period. The data for
the stock returns of the companies and information on the stock indices is obtained from the OECD
database, Yahoo finance and NASDAQ composite. Only those stock indices and variables have
been included in the study that have sufficient amount of information available. Therefore, several
variables were excluded from the study that did not have sufficient amount of information was not
available.
Variables
Dependent variables
For the subject of econometrics, the dependent variables are those variables whose movements are
dependent or evaluated by the changes in the other variables also known as dummy and/or
variables (Korner and Wahlgren, 2006).
Share Prices
Share price is the dependent variable that will be used in this research paper. The share prices
obtained for the study will be obtained from the online electronic sources. The credibility of the
sources will be mentioned in the research paper. The stock prices will offer information on the
fluctuations of the stock market of the US. Rather than making use of the index value, the monthly
returns of the US stock market will be calculated. The reasoning behind using the monthly returns
rather than index data values is that monthly returns will produce more reliable regression results
because of the fact that the variables might correlate otherwise due to the trend. The method will
be used for both oil prices and GDP but the interest rates will be kept unchanged. The reason for
this is the use of the interest rates in the percentage form rather than money and thus they will not
be affected by the trend changes.
Date
Share Prices
1/1/2012
106.8289
2/1/2012
111.7057
3/1/2012
112.9959
4/1/2012
111.3108
5/1/2012
106.6202
6/1/2012
104.5375
7/1/2012
107.5475
8/1/2012
110.8503
9/1/2012
114.3767
10/1/2012
114.6529
11/1/2012
112.475
12/1/2012
115.8117
1/1/2013
121.032
2/1/2013
123.1245
3/1/2013
125.1001
4/1/2013
125.8008
5/1/2013
130.6188
6/1/2013
127.3489
7/1/2013
130.7415
8/1/2013
131.3965
9/1/2013
133.128
10/1/2013
136.161
11/1/2013
139.8263
12/1/2013
140.4791
1/1/2014
141.7957
2/1/2014
140.883
3/1/2014
144.0925
4/1/2014
145.449
5/1/2014
147.3277
6/1/2014
150.8834
7/1/2014
151.9819
8/1/2014
149.9889
9/1/2014
151.2588
10/1/2014
144.883
11/1/2014
151.0701
12/1/2014
149.7646
1/1/2015
147.8479
2/1/2015
151.8315
3/1/2015
151.0605
4/1/2015
153.417
5/1/2015
154.2413
6/1/2015
152.4042
7/1/2015
150.2334
8/1/2015
145.7222
9/1/2015
137.721
10/1/2015
142.8564
11/1/2015
144.2752
12/1/2015
141.1625
1/1/2016
131.7877
2/1/2016
130.3594
3/1/2016
139.1554
4/1/2016
143.0591
5/1/2016
143.0577
6/1/2016
144.0741
7/1/2016
147.7299
8/1/2016
149.2793
9/1/2016
148.0711
10/1/2016
146.3304
11/1/2016
147.5157
12/1/2016
153.4093
1/1/2017
155.1524
2/1/2017
158.0345
3/1/2017
159.2511
4/1/2017
158.4775
5/1/2017
160.1226
6/1/2017
162.3951
7/1/2017
164.2544
Mean
138.4796866
Median
143.0591
Independent variables
In the regression analysis, an independent variable is a variable used with along with the more
independent variables to have an impact on the dependent variables (Korner and Wahlgren, 2006).
Therefore, in this research paper, all of the independent variables are supposed to have an impact
on the share price.
Long-term interest rate
The long-term interest rates are applicable for 10 years. When the long-term interest is low, the
stock market and the country can be perceived as stable and safe. This would result in an increase
in the investments in the stock market. Conversely, an opposite effect will be witnessed if there is
a major increase in the long-term interest rate. Thereby, the investors will not be willing to invest
in the market.
Date
Long term interest rate
1/1/2012
1.97
2/1/2012
1.97
3/1/2012
2.17
4/1/2012
2.05
5/1/2012
1.8
6/1/2012
1.62
7/1/2012
1.53
8/1/2012
1.68
9/1/2012
1.72
10/1/2012
1.75
11/1/2012
1.65
12/1/2012
1.72
1/1/2013
1.91
2/1/2013
1.98
3/1/2013
1.96
4/1/2013
1.76
5/1/2013
1.93
6/1/2013
2.3
7/1/2013
2.58
8/1/2013
2.74
9/1/2013
2.81
10/1/2013
2.62
11/1/2013
2.72
12/1/2013
2.9
1/1/2014
2.86
2/1/2014
2.71
3/1/2014
2.72
4/1/2014
2.71
5/1/2014
2.56
6/1/2014
2.6
7/1/2014
2.54
8/1/2014
2.42
9/1/2014
2.53
10/1/2014
2.3
11/1/2014
2.33
12/1/2014
2.21
1/1/2015
1.88
2/1/2015
1.98
3/1/2015
2.04
4/1/2015
1.94
5/1/2015
2.2
6/1/2015
2.36
7/1/2015
2.32
8/1/2015
2.17
9/1/2015
2.17
10/1/2015
2.07
11/1/2015
2.26
12/1/2015
2.24
1/1/2016
2.09
2/1/2016
1.78
3/1/2016
1.89
4/1/2016
1.81
5/1/2016
1.81
6/1/2016
1.64
7/1/2016
1.5
8/1/2016
1.56
9/1/2016
1.63
10/1/2016
1.76
11/1/2016
2.14
12/1/2016
2.49
1/1/2017
2.43
2/1/2017
2.42
3/1/2017
2.48
4/1/2017
2.3
5/1/2017
2.3
6/1/2017
2.19
7/1/2017
2.32
Mean
2.1567164
Median
2.17
Short-term interest rate
The second independent variable to be used in the study is the short-term interest rate. The short-
term interest rate variable will help in the explanation of the share prices. This is due to the fact
that when there is a fall in the short-term interest rate, the public gets many benefits and is favored
by them. Due to which, the companies will increase their lending capabilities and make higher
investments. In addition, people will avoid keeping their money in the bank as it will not be
profitable in comparison which will lead to higher liquidity in the stock market. Conversely, an
opposite effect will be witnessed if there is an increase in the short-term interest rate.
Date
Short term interest rate
1/1/2012
0.4
2/1/2012
0.3
3/1/2012
0.29
4/1/2012
0.29
5/1/2012
0.29
6/1/2012
0.32
7/1/2012
0.3
8/1/2012
0.26
9/1/2012
0.24
10/1/2012
0.23
11/1/2012
0.23
12/1/2012
0.24
1/1/2013
0.23
2/1/2013
0.22
3/1/2013
0.21
4/1/2013
0.2
5/1/2013
0.2
6/1/2013
0.19
7/1/2013
0.14
8/1/2013
0.12
9/1/2013
0.11
10/1/2013
0.12
11/1/2013
0.12
12/1/2013
0.14
1/1/2014
0.12
2/1/2014
0.13
3/1/2014
0.12
4/1/2014
0.12
5/1/2014
0.11
6/1/2014
0.11
7/1/2014
0.13
8/1/2014
0.13
9/1/2014
0.12
10/1/2014
0.12
11/1/2014
0.13
12/1/2014
0.15
1/1/2015
0.16
2/1/2015
0.15
3/1/2015
0.14
4/1/2015
0.13
5/1/2015
0.15
6/1/2015
0.18
7/1/2015
0.19
8/1/2015
0.26
9/1/2015
0.27
10/1/2015
0.25
11/1/2015
0.3
12/1/2015
0.54
1/1/2016
0.57
2/1/2016
0.54
3/1/2016
0.55
4/1/2016
0.55
5/1/2016
0.57
6/1/2016
0.55
7/1/2016
0.62
8/1/2016
0.73
9/1/2016
0.75
10/1/2016
0.72
11/1/2016
0.71
12/1/2016
0.87
1/1/2017
0.9
2/1/2017
0.87
3/1/2017
0.98
4/1/2017
1.03
5/1/2017
1.05
6/1/2017
1.16
7/1/2017
1.22
Mean
0.3662687
Median
0.24
GDP
The gross domestic product (GDP) is considered to be an indicator to gauge the economic health
of a country. The GDP is used to assess all the economic activities of a country during a particular
time period, generally quarterly or yearly. All of the variables mentioned above are quite simple
and straightforward to access and procure but GDP is a bit tricky. The reason is that GDP is usually
measured annually, it will be problematic for getting monthly GDP which is required in the
research.
Date
GDP
1/1/2012
15.86
2/1/2012
16.07
3/1/2012
15.98
4/1/2012
16.09
5/1/2012
16.12
6/1/2012
16.15
7/1/2012
16.24
8/1/2012
16.16
9/1/2012
16.28
10/1/2012
16.21
11/1/2012
16.27
12/1/2012
16.41
1/1/2013
16.5
2/1/2013
16.43
3/1/2013
16.49
4/1/2013
16.52
5/1/2013
16.46
6/1/2013
16.64
7/1/2013
16.69
8/1/2013
16.78
9/1/2013
16.78
10/1/2013
16.88
11/1/2013
17.05
12/1/2013
17.06
1/1/2014
16.95
2/1/2014
17.02
3/1/2014
17.1
4/1/2014
17.17
5/1/2014
17.31
6/1/2014
17.37
7/1/2014
17.49
8/1/2014
17.64
9/1/2014
17.58
10/1/2014
17.64
11/1/2014
17.74
12/1/2014
17.7
1/1/2015
17.7
2/1/2015
17.86
3/1/2015
17.79
4/1/2015
17.91
5/1/2015
17.99
6/1/2015
18.1
7/1/2015
18.04
8/1/2015
18.12
9/1/2015
18.26
10/1/2015
18.18
11/1/2015
18.23
12/1/2015
18.26
1/1/2016
18.28
2/1/2016
18.2
3/1/2016
18.36
4/1/2016
18.44
5/1/2016
18.42
6/1/2016
18.49
7/1/2016
18.57
8/1/2016
18.68
9/1/2016
18.78
10/1/2016
18.72
11/1/2016
18.91
12/1/2016
18.97
1/1/2017
18.94
2/1/2017
19.04
3/1/2017
19.1
4/1/2017
19.04
5/1/2017
19.24
6/1/2017
19.32
7/1/2017
19.71
Mean
17.559403
Median
17.64
Dummy Variable
A dummy variable in regression is that variable which in the natural state does not include numbers
wherein its several parts need to be arranged in an apt order so that they fit in any econometric
model. It is due to the dummy variables only that the quantitative variables are fit into the analysis
(Dougherty, 2011). The method to do so is to assign a value (0 or 1) to its parts. For a certain year,
a dummy variable will be assigned a value 1 if the encoded part is in compliance with the natural
variable and conversely, it is assigned the value 0. For this study, election date is the dummy
variable. The time period before the election result is assigned the value “0” and the time period
after the election result is assigned the value “1”.
Analysis and Discussion
The results of the regression output are shown in the tables below. Initially, the focus of the study
is kept on the p-values of the dummy variables. This is because the p-values will impart
information on the relationship between the performances of the stock market and share prices,
and the 2016 US Presidential election. The p-values help in the evaluation of the relationship
between the independent variables and the dependent variables. This is achieved by reviewing the
coefficients. In the table below, the results can be seen for the share prices along with the adjusted
R square. The high values of R square and adjusted R square reveal that the model fits the data
efficiently as it is able to explain more than 90% of the overall variations in the sample data above
average.
The return values of the variables is calculated by:
Step 1: All the data (prices) are converted to the natural logs as follows:
LSHARE= LOG (SHARE_PRICES)
LLONG= LOG (LONG_TERM)
LSHORT= LOG (SHORT_TERM)
LGDP= LOG (GDP)
Step Two: The returns of the above variables are found as follows:
RSHARE= LSHARE - LSHARE (-1)
RLONG= LLONG – LLONG (-1)
RSHORT= LSHORT – LSHORT (-1)
LGDP= LGDP – LGDP (-1)
Then, the regression output is obtained from Eviews. The table below reveals the results when the
returns on the RSHARE_PRICES is taken as the dependent variable in the regression analysis.
In the test results, the coefficients of the short term interest rate are negative that means that it has
a negative relationship with the dependent variable. Therefore, higher the values for
RSHORT_TERM, lower will be the values of RSHARE_PRICES. The sign should have been
positive as an increase in RSHORT_TERM, leads to increase in RSHARE_PRICES.
The test results reveal that the p-values are more than 0.05, the test indicates that all of the
explanatory variables have not been chosen appropriately as they suit the model.
The coefficient of the dummy variable is 0.005632, which is positive. This reveals that there has
been a change in the RSHARE_PRICE, pre and post 2016 US election result. But the p value is
greater than 0.05 which means that the impact of the dummy on RSHARE_PRICE is not positive.
The RLONG_TERM and RGDP have coefficients as 0.098109 and 0.380650, which means that
the relationship between the variables and RSHARE_PRICE is positive. This means that the one
unit increase in the variables will lead to an 8.634026 and 17.16016 units increase in
RSHARE_PRICES. The RSHORT_TERM is revealed to show a negative relationship with the
RSHARE_PRICES.
The R square value of the test results is 0.209147 that means that the explanatory variables are not
effective in explaining the changes in RSHARE_PRICES. The p-values of the t-statistic of
RLONG_TERM and RSHORT_TERM are less than 0.05 which means that the variables impact
the RSHARE_PRICES significantly. The Durbin-Watson value of the result is 1.911070 that is
used to check if there is an auto-serial correlation present in the data. The value lies on the left of
the (1, 2, 3, 4) scale that means that there is a negative auto serial correlation present.
ADF Testing
The test results and outputs are included in the appendix. The testing check each and every variable
for stationarity. We make use of the market model as discussed in the methodology section above.
The return for the US share prices SHARE_PRICES at time t is:
RSHARE_PRICES = β0 + β1*RLONG_TERM + β2*RSHORT_TERM + β3*RGDP +
β4*DUMMY
Where, RSHARE_PRICES, RLONG_TERM, RSHORT_TERM and RGDP are the returns on
the US Share Prices, Long term interest rates, Short term interest rates and GDP respectively. We
run the above model to find the two parameters, α and β by using OLS.
We use the estimation period data. We need to identify whether the series are stationary or not
before running a regression between the series. We make use of the return values of all the
variables as shown above in Step 1 and Step 2 and the prices of the above two variables are
transferred to Eviews.
The First Series: RSHARE
The formal approach has been employed to determine the stationarity of the series. For the ADF
test, we formulate two hypotheses; null hypothesis (H0) and alternative hypothesis (H1).
1. H0: series has a unit root and it is not stationary.
2. H1: series has no unit root and it is stationary.
The value of α is taken to be 5% or 0.05 against which the p values of the test are compared. Refer
to the Table for the ADF test results for RSHARE. As is evident from the Table 1, the value for
the test statistic is -7.072866. The p value for the variable as obtained from the ADF test is 0.0000.
This p vale is lesser than 0.05, therefore the null hypothesis (H0) is rejected and alternative
hypothesis (H1) is accepted. This means that the RSHARE series has no unit root and it is
stationary. This means that the variance and mean of the RSHARE time series is constant over the
period Jan 2012-Aug 2017. The forecasting for a stationary time series is easy through
mathematical transformations.
The Second Series: RLONG
For this ADF tesing too, formal approach was employed for determining the stationarity of the
RLONG series. For the ADF test, we formulate two hypotheses; null hypothesis (H0) and
alternative hypothesis (H1).
1. H0: series has a unit root and it is not stationary.
2. H1: series has no unit root and it is stationary.
The value of α is taken to be 5% or 0.05 against which the p values of the test are compared. Refer
to the Table 2 for the ADF test results for RLONG. As is evident from the Table 2, the value for
the test statistic is -6.146152. The p value for the variable as obtained from the ADF test is 0.0000.
This p vale is lesser than 0.05, therefore the null hypothesis (H0) is rejected and alternative
hypothesis (H1) is accepted. This means that the RLONG series has no unit root and it is stationary.
This means that the variance and mean of the RLONG time series is also constant over the period
Jan 2012-Aug 2017.
The Third Series: RSHORT
For determining the stationarity of the series, two hypotheses were formulated, namely, null
hypothesis (H0) and alternative hypothesis (H1).
1. H0: series has a unit root and it is not stationary.
2. H1: series has no unit root and it is stationary.
The value of α is taken to be 5% or 0.05 against which the p values of the test is compared. Refer
to the Table 3 for the ADF test results for RSHORT. As is evident from the Table 3, the value for
the test statistic is -6.471967. The p value for the variable as obtained from the ADF test is 0.0000.
This p vale is lesser than 0.05, therefore the null hypothesis (H0) is rejected and alternative
hypothesis (H1) is accepted. This means that the RSHORT series has no unit root and it is
stationary. This means that the variance and mean of the RSHORT time series is constant over the
period Jan 2012-Aug 2017.
The Fourth Variable: RGDP
In the ADF test of the RGDP time series to check whether the series is stationary or not, two
hypotheses were formulated. These two hypothesis are null hypothesis (H0) and alternative
hypothesis (H1).
1. H0: series has a unit root and it is not stationary.
2. H1: series has no unit root and it is stationary.
The value of α is taken to be 5% or 0.05 against which the p values of the test are compared.
Refer to the Table 4 (Appendix) for the ADF test results for RGDP. As is evident from the Table
4, the value for the test statistic is -10.68759. The p value for the variable as obtained from the
ADF test is 0.0000. This p vale is lesser than 0.05, therefore the null hypothesis (H0) is rejected
and alternative hypothesis (H1) is accepted. This means that the RGDP series has no unit root
and it is stationary. This means that the variance and mean of the RGDP time series is constant
over the period Jan 2012-Aug 2017.
In addition all the four series have been shown in the graphs. Refer to the appendix Table 5,
Table 6, Table 7 and Table 8 that show the graphs of the RSHARE, RLONG, RSHORT and
RGDP time series. The graph reveal that all of the four time series are stationary and I(0).
Correlogram Analysis
The Correlogram in Eviews offer the correlation statistics for a time series. For the analysis, it
would plot the sample autocorrelations and the time lags. The correlogram analysis has been
done for all the four time series and results are attached in the appendix (Table 9, Table 10, Table
11 and Table 12).
For all of the four time series, two hypothesis were formulated. These two hypothesis are null
hypothesis (H0) and alternative hypothesis (H1).
1. H0: The series has no unit root.
2. H1: The series has a unit root.
The value of α was taken to be 5% or 0.05 against which the Prob. Value are compared.
RSHARE time series: In this test series, there are no test statistics to calculate as seen in the Table
9 (Refer to Appendix). Therefore, the Prob column is considered for analysis. On analysis, it is
observed that all of the p values are higher than 0.05. As p-values are higher than 0.05, we cannot
reject the H0. Thus, we accept the H0 that states that the series has no unit root. This means that
the RSHARE time series is stationary.
RLONG time series: In this case, there are no test statistics to calculate as seen in the Table 10
(Refer to Appendix). Therefore, the Prob column is considered for analysis. It is observed that all
of the p values are higher than 0.05. So, we cannot reject the null hypothesis (H0). Thus, we accept
H0 that states that the series has no unit root. This means that the RLONG time series is stationary.
RSHORT time series: There are no test statistics in the results as evident from the Table 11 in the
appendix. Therefore, the Prob column is considered for analysis. On analysis, it is observed that
all of the p values are higher than 0.05. As p-values are higher than 0.05, we cannot reject the null
hypothesis (H0). Thus, we accept the H0 that states that the series has no unit root. This means
that the RSHORT time series is stationary.
RGDP time series: For RGDP test series too, there are no test statistics to calculate as seen in the
appendix (Refer to Table 12). Therefore, the Prob column is considered for analysis purposes. On
analysis, it is observed that all of the p values are higher than 0.05. As p-values are higher than
0.05, we cannot reject the null hypothesis. Thus, we accept the H0 that states that the series has no
unit root. This means that the RGDP time series is stationary.
Cointegration Test
For cointegration test, we observe the regression output of the variables (Refer to Appendix: Table
13). From this output, we obtain a graph of the residuals (Refer to Appendix: Table 15). The graph
shows that the residual is stationary. After this, we do the correlogram analysis and for this purpose
we formulate two hypothesis:
1. H0 : The series has no unit root
2. Ha : The series has a unit root
3. α = 5% or 0.05
We obtain the correlogram test results at first difference for the residual. Refer to the Table 14 in
the appendix for the test results. For analysis purposes, there are no test statistics to calculate
therefore, we look at the Prob column. It is observed that all of the p values are higher than 0.05.
As p-values are > 0.05, we cannot reject the H0. So, we accept the H0 that state that the series
has no unit root, meaning that the series is stationary.
The unit root test of the residual
The unit root test of the residual checks whether the residual time series is stationary or not and
doesn’t have a unit root. For the unit root test of the residual, we developed two hypotheses:
1. H0: series has a unit root and it is not stationary
2. H1: series has no unit root and it is stationary.
3. α = 5% or 0.05
Then, we perform the ADF testing at first difference. As seen from the ADF test results, the
value of the test statistic is -8.816275 (Refer to Appendix: Table 16). The p value of the test is
0.0000 that is less than 0.05. Therefore, we reject H0 hypothesis and accept H1 that state that the
series has no unit root and it is stationary.
The Regression Line Using the Estimated Period
The regression output of the variables can be seen from Table 13 (Refer to Appendix). The
equations that are developed from the test results are:
RSHARE_PRICE = C(1) + C(2)*RLONG_TERM + C(3)*RSHORT_TERM + C(4)*RGDP +
C(5)*DUMMY
RSHARE_PRICE = 0.00516009940916 + 0.0981087678167*RLONG_TERM-
0.0536538599523*RSHORT_TERM + 0.380649511019*RGDP +
0.00563155137041*DUMMY
Now, the regression is tested for returns to check if there are abnormal returns or not. For this
purpose, we formulate two hypotheses:
1. H0: No abnormal return
2. Ha: There is abnormal return
3. α = 5% or 0.05
In order to calculate the test statistic (t), we make use of the formula t= CAR/ (σ
CAR
/ √N). We
use the calculated cumulative abnormal return (CAR) found for the event day (November 2017)
and for the denominator and we use the value of the standard error of the regression (the
regression of the estimated window) for the denominator (Refer to Appendix: Table 13). The
value of the t statistic comes out to be -1.16. For calculations, refer to Table 17 in the Appendix.
In this case, the probability is between 0.01 and 0.02. Both are smaller than 0.05. Because they
are in the rejection area; we reject the null hypothesis. Therefore, we can conclude that the
abnormal return are present for the output.
The regression analysis for the returns was used to calculate the abnormal returns for the variables.
The equation for the share prices returns came out to be:
RSHARE_PRICE= 0.005160 + 0.098109*RLONG_TERM – 0.053654*RSHORT_TERM
+0.380650*RGDP + 0.005632*DUMMY
The value of α is 0.005160 and for β1, β2 and β3 are 0.098109, -0.053654 and 0.380650
respectively. We have returns for all of the variables. The estimated abnormal returns are
calculated with the help of the formula:
AR= RSHARE_PRICES- α - β1(RLONG_TERM)- β2(RSHORT_TERM)- β3(RGDP)
Wherein, RSHARE_PRICES, RLONG_TERM, RSHORT_TERM, RGDP are the returns for the
variable. The next step involves the aggregation of the ARs to find the CAR (cumulative Abnormal
Return). The abnormal returns can be seen in the Table 18 (Refer to Appendix).
The test results showed significant variations based on the previous research conducted in this
field. Majority of the scientific articles maintain that the stock prices increases for the period
surrounding the election results while some reveal that there is no correlation between the election
and the stock markets whatsoever (Bialkowski et al, 2006). In this research, it was found out that
there is an impact of the US elections on the stock market performances. The dummy variable in
the regression analysis had two values: before election (0) and after election (0). The results
showed that the coefficient for the dummy variable was 8.662282. The value is very high which
means that there have been significant variations before and after 2016 US elections. The R square
and adjusted R square values were at 90% that shows how effectively the explanatory variables
explain the model with a good fit regression.
Prior to the election results, the share prices are seen to be declining rapidly. The share prices
started to grow for October, i.e. one month before the election and kept on growing exponentially.
For the year 2017, it reached its highest value of 164. The US elections are known to be world
events and such events are known to impact the stock market returns. The uncertainty surrounding
such events impacts the investing behavior of the traders. The US stock market was projected to
show improved performances during and after 2016 US elections as Trump increased the
uncertainty factor (Granville, K., 2016). The DJIA had the best figure since 2011, the next day
after the election. Many stock indices showed a dip on the announcement of the election results.
But, the indices soon recovered from the fall. It is not possible to develop a single theory that can
explain this phenomenon. All the test results show that the elections have significant impact on the
performances of the stock market. Therefore, both the hypotheses formulated for the research were
proven true as there is an impact of elections on the stock market and the US stock market showed
improved performances after the Trump victory.
Conclusion and Future Research
Political impacts on the share trading market are verifiably considered to be marginal. The stock
market variations that surround the US presidential elections reveal that the election years show a
different performance level in comparison to the non-election years. Therefore, the impact of the
political influences and elections on the stock market cannot be neglected.
This research paper studied the impact of the 2016 US Presidential election on the stock market
and evaluated whether it was distinct in the cases of election years and non-election years. While
some of the researchers have conducted study on this subject previously, generally used the results
of the study to predict the Presidents rather than determining the impact on the stock markets. Also,
there were not many US-based study conducted in this field. The 2016 US election is a unique
opportunity for research purposes as both of the candidates were controversial. Hillary Clinton
could have been the first female President and was involved in an illegal email scam. On the other
hand, Donald Trump was experienced in entertainment and business but had no political
experience whatsoever. He also attracted attention with his sexist and racist statements.
The impact of the US election on the stock market was evaluated by determining the volatility of
the share prices. These are calculated with the help of the prices of the company shares that are
traded on the foreign or national stock exchanges. Share prices are determined through stock
exchange with the help of monthly data that is expressed in terms of arithmetic average of daily
data. Other data that was required was related to long term interest rate, short term interest rate and
GDP. After the data was collected, a regression analysis was performed. Prior studies dictate that
variation in the government have a negative impact on the US stocks. However, the impact of the
Trump win on the stock market was highly unanticipated. Generally, companies make choices on
the basis of the projected future economic decisions of the winning administration (Schiereck et
al., 2016). From this perspective, the Trump win can be seen as a major change in the government
policy. Usually, such changes are associated with a decline in the stock prices and the dip is greater
if the uncertainty is higher (Pastor and Veronesi, 2012). When the uncertainty gets mitigated, the
stock market recovers (Pantzalis et. al., 2000). For US 2016 elections, traders and investors saw a
Trump win as a risk to the market but soon embraced the results. The stock tumbled initially but
the Trump promise of rolling back regulations, cutting taxes and improving infrastructure made
the investors believe in Trump victory and the stock prices witnessed growth. The Trump’s
approach of America First favoring isolationism encouraged investments on national and foreign
markets resulted in positive impact on the stocks. It is known that all the US companies faced a
political uncertainty in relation to the 2016 elections and variation in the results were observed.
A future research can be conducted to determine whether there is any correlation between the US
Presidential elections and the stock markets of other countries since this study has only focused on
the stock market of the United States. Another subject for research would be to determine if the
political inclination of the incumbent party and the other party impact the stock market of the
country. This research may take help of the theories of the behavioral sciences.
References
"Key Trends That Shaped The Markets In The First Half Of 2016 - Www.Dealogic.Com".
www.dealogic.com. N.p., 2017. Web. 4 August 2017.
Alesina, A. (1987). Macroeconomic policy in a two-party system as a repeated game. Quarterly
Journal of Economics, (102): 651–678.
Allvine, F. C., and O’Neil, D. E. (1980). Stock market returns and the presidential election cycle:
Implications for market efficiency. Financial Analysts Journal, 36, pp. 49-56.
Allvine, F. C., and O'Neill, D. (1980). Stock Market Returns and the Presidential Election
Cycle/Implications for Market Efficiency. Financial Analysts Journal, 36(5): 49–56.
Analysts Journal, 48(3), pp. 5-8. Print.
Bartholdy, J., Olson, D., and Peare, P. (2006). Conducting Event Studies on Small Stock
Exchanges. The European Journal of Finance, Volume 13, Issue 3 (2007), pp. 227-252.
Bialkowski, Jedrzej Pawel, Katrin Gottschalk, and Tomasz Piotr Wisniewski. "Stock Market
Bialkowski, Jedrzej Pawel, Katrin Gottschalk, and Tomasz Piotr Wisniewski. (2006). "Stock
Market Volatility Around National Elections". SSRN Electronic Journal n. pag. Web.
Binder, J. (1998). The Event Study Methodology Since 1969. Review of Quantitative Finance and
Accounting, Volume 11, Issue 2 (Sep. 1998), pp.111-137.
Bohl, M. T. and Gottschalk, K. (2006). International evidence on the Democrat premium and the
presidential cycle effect. The North American Journal of Economics and Finance, 17(2): pp. 107–
120.
Bohl, M.T., and Gottschalk, K. (2004). International evidence on the democratic premium and the
presidential cycle effect. Working Paper.
Booth, J. and Booth, L. (2003). Is presidential cycle in security returns merely a reflection of
business conditions? Review of Financial Economics, 12, pp. 131-159.
Booth, J. R., and Booth, L. C. (2003). Is presidential cycle in security returns merely a refl ection
of business conditions? Review of Financial Economics, 12 (2): 131-159.
Brown, K.C., Harlow, W.V., and Tinic, S.M. (1988). Risk aversion, uncertain information, and
market efficiency. Journal of Financial Economics, 22, 355–385.
Brown, S., and Warner, J. (1980). Measuring Security Price Performance. Journal of Financial
Economics, Volume 8, Issue 3 (Apr. 1980), pp. 205-258.
Brown, S., and Weinstein M. (1985). Derived Factors in Event Studies. Journal of Financial
Economics, 14, 491-495.
Caporale, T., and Grier, K. (2000). Political Regime Change and the Real Interest Rate. Journal
of Money, Credit and Banking, 32(3): 320–34.
Caro, R. (2002). The years of Lyndon Johnson: Master of the Senate. New York, NY: Alfred
Knopf.
Corrado. (2011), Event Studies: A Methodology Review. Accounting & Finance, Volume 51,
Issue 1 (Mar. 2011), pp. 207-234.
Dopke, J. and Pierdzioch, C. (2006). Politics and the stock market: Evidence from Germany.
European Journal of Political Economy, 22(4): pp. 925–943.
Döpke, J., and Pierdzioch, C. (2006). Politics and the stock market: Evidence from Germany.
European Journal of Political Economy, 22 (4): 925-943.
Dougherty, C. (2011). Introduction to Econometrics, 4th Edition, Oxford University Press.
Downs, A. (1957). An Economic Theory of Democracy. New York: Harper.
Fama, E. (1970). Efficient capital markets: A review of theory and empirical work. Journal of
Finance, 25, 383-417.
Financial Market Perspective. American Journal of Political Science, 34(2): 289–310.
Fiorina, Morris P. (1991). Elections and the Economy in the 1980s: Short- and Long-Term Effects.
In Politics and Economics in the Eighties, eds. Alberto Alesina and Geoffrey Carliner. NBER
Books, pp. 17–40.
Gärtner, Manfred, and Wellershoff, K. (1995). Is There an Election Cycle in American Stock
Returns? International Review of Economics and Finance, 4(4): 387–410.
Gerber, Alan S., Huber, G. and Washington, E. (2009). Party Affiliation, Partisanship, and Political
Beliefs: A Field Experiment. The National Bureau of Economic Research (NBER). Working Paper
15365.
Goodell, J. W., and Bodey, R. A. (2012). Price-earnings changes during US presidential election
cycles: Voter uncertainty and other determinants. Public Choice, 150, pp. 633-650.
Goodell, J. W., and Vähämaa, S. (2013). U.S. presidential elections and implied volatility: The
role of political uncertainty. Journal of Banking & Finance, 37, pp. 1108–1117.
Granville, K., (2016). This Was the Trans-Pacific Partnership, NY Times,
http://www.nytimes.com/interactive/2016/business/tpp-explained-what-is-trans-pacific-partnersh
ip.html?_r=0 (retrieved 2017-08-23)
He, Y., Hai, L. and Chunchi, W. (2009). The 2000 Presidential Election and the Information Cost
of Sensitive Versus Non-sensitive S&P500 Stocks. Journal of Financial Markets, 12(1), pp. 54–
86.
Hensel, C. and Ziemba W. (1995). United States investment returns during democratic and
republican administrations, 1928-1993. Financial Analysts Journal March-April: 61-69.
Herbst A. and Slinkman C. (1984). Political-Economic Cycles in the U.S. Stock Market. Financial
Analysts Journal, March-April, 38-44.
Herbst, A. F., and Slinkman, C. W. (1984). Political-Economic Cycles in the U.S. Stock Market.
Financial Analysts Journal, 40(2): 38–45.
Herron, Michael C., James, L., Donald, C., and Silver, J. (1999). Measurement of Political Effects
in the United States Economy: A Study of the 1992 Presidential Election. Economics and Politics
11(1): 51–81.
Hibbs, D. (1977). Political parties and macroeconomic policy. American Political Science Review,
(71): 1467-87.
Hong, H. and Kostovetsky, L. (2012). Red And Blue Investing: Values And Finance. Journal of
Financial Economics, 103.1 (2012): 1-19. Web.
Huang, Roger D. (1985). Common Stock Returns and Presidential Elections. Financial Analysts
Jacobsen, D.I. (2002). Vad hur och varför: om metodval i företagsekonomi och andra
samhällsvetenskapliga ämnen, Lund: Studentlitteratur AB.
Johnson, R. R., Chittenden, W. and Jensen, G. (1999). Presidential Politics, Stocks, Bonds, Bills,
and Inflation. Journal of Portfolio Management, 26(1), pp. 27 – 31.
Jones, Randall J. (2008). The State of Presidential Election Forecasting: The 2004 Experience.
International Journal of Forecasting, 24(2), pp. 310–21.
Journal, 41(2): 58–61.
Julio, B. and Yook, Y. (2012). Political Uncertainty And Corporate Investment Cycles. SSRN
Electronic Journal n. pag. Web.
Kitchin, J. (1923). Cycles and trends in economic factors. Review of Economics and Statistics,
5(1), 10-17.
Körner, S. and Wahlgren, L. (2006). Statistisk Dataanalys, Lund: Studentlitteratur AB.
Körner, S. and Wahlgren, L. (2006). Statistisk Dataanalys, Lund: Studentlitteratur AB.
Körner, S. and Wahlgren, L. (2013). Praktisk Statistik, Lund: Studentlitteratur AB.
Leblang, D. and Mukherjee, B. (2005). Partisan politics, interest rates and the stock market:
evidence from American and British returns in the twentieth century. Economics and Politics, (19):
135-167.
Lo, Andrew W., and McKinley, C. (1990). An Econometric Analysis of Nonsynchronous Trading.
Journal of Econometrics, Volume 45, Issue 1 (Jul./Aug. 1990), pp. 181-211.
MAcKinlay, A. (1997). Event Studies in Economics and Finance. Journal of Economic Literature,
Vol. 35, pp. 5-101.
Pantzalis, C., Stangeland, D.A. and Turtle, H.J. (2000). Political elections and the resolution of
uncertainty: the international evidence. J. Banking Finance, 24 (10), 1575–1604.
Pástor, L., and Veronesi, P. (2012). Uncertainty about government policy and stock prices. J.
Finance 67 (4), 1219–1264.
Riley, W. B., and Luksetich, W.A. (1980). The market prefers Republicans: Myth or reality.
Journal of Financial and Quantitative Analysis, 15, 541-560.
Roberts, Brian E. (1990). Political Institutions, Policy Expectations, and the 1980 Election: A
S.V.D.Nageswara Rao, and Sreejith, U. (2014). Event Study Methodology: A Critical Review, The
Macrotheme Review, Volume 3, Issue 1 (Spring 2014), pp. 40-53.
Saadi, S., Devinder G., and Khaled, E. (2006). On the Validity of Conventional Statistical Tests
Given Evidence of Non-Synchronous Trading and Non-Linear Dynamics in Returns Generating
Process. Applied Economics Letters, Volume 13, Issue 5 (Aug. 2006), pp. 301-305.
Santa-Clara, P., and Valkanov R. (2003). The presidential puzzle: Political cycles and the stock
market. The Journal of Finance, 58, pp. 1841-1872.
Schiereck, D., Florian K., and Kolaric, S. (2016). Brexit: (Not) another Lehman moment for
banks? Finance Research Letters (in press).
Shalit, H., and Shlomo, Y. (2002). Estimating Beta. Review of Quantitative Finance and
Accounting, Volume 18, Issue 2 (Mar. 2002) 95-118.
Shleifer A. (2000). Do Demand Curves for Stocks Slope Down. Journal of Finance, 41, 579 590.
Snowberg, E, Wolfers, J. and Zitzewitz, E. (2007). Partisan Impacts on the Economy: Evidence
from Prediction Markets and Close Elections. The Quarterly Journal of Economics, 122(2): 807–
29.
Stovall, R. (1992). Forecasting Stock Market Performance Via The Presidential Cycle. Financial
Sweet, Ryan, Ozimek, A. and Asher, K. (2016). Economics Of U.S. Policy Uncertainty. Moody's
Analytics: Regional Financial Review, n. page. Print.
Volatility Around National Elections". SSRN Electronic Journal n. pag. Web.
Wisniewski, T. (2009). Can Political Factors Explain The Behavior Of Stock Prices Beyond The
Standard Present Value Models?. SSRN Electronic Journal, n. pag. Web.
Wong, W. and McAleer, M. (2009). Financial Astrology: Mapping The Presidential Election Cycle
In US Stock Markets. SSRN Electronic Journal n. pag. Web.
Wong, W.-K., and McAleer, M. (2009). Mapping the Presidential Election Cycle in US stock
markets. Mathematics and Computers in Simulation, 79 (11): 3,267–3,277.
Appendix
1. ADF Test RSHARE
2. ADF test RLONG
3. ADF test RSHORT
4. ADF test RGDP
5. RSHARE series graph
-.08
-.06
-.04
-.02
.00
.02
.04
.06
.08
I II III IV I II III IV I II III IV I II III IV I II III IV I II III
2012 2013 2014 2015 2016 2017
RSHARE
6. RLONG series Graph
-.20
-.15
-.10
-.05
.00
.05
.10
.15
.20
I II III IV I II III IV I II III IV I II III IV I II III IV I II III
2012 2013 2014 2015 2016 2017
RLONG
7. RSHORT series Graph
-.4
-.2
.0
.2
.4
.6
I II III IV I II III IV I II III IV I II III IV I II III IV I II III
2012 2013 2014 2015 2016 2017
RSHORT
8. RGDP series Graph
-.010
-.005
.000
.005
.010
.015
.020
.025
I II III IV I II III IV I II III IV I II III IV I II III IV I II III
2012 2013 2014 2015 2016 2017
RGDP
9. Correlogram RSHARE
10. Correlogram RLONG
11. Correlogram RSHORT
12. Correlogram RGDP
13. Regression Output
14. Correlogram test of the residual
15. Graph of the residual
-.08
-.06
-.04
-.02
.00
.02
.04
.06
I II III IV I II III IV I II III IV I II III IV I II III IV I II III
2012 2013 2014 2015 2016 2017
RSHARE_PRICE Residuals
16. ADF Test- Unit root of regression output
17. The test statistic is: t= CAR/ (σ
CAR
/ √N)
= -0.02602/0.022442= -1.16
18. Abnormal Returns and CAR
RSHARE_PRIC
E
RLONG_TER
M
RSHORT_TER
M
RGDP
AR
CAR
-
0.00516
-
0.00516
0.044639245
0
-0.287682072
0.01315396
4
0.01903
7
0.01387
7
0.011483801
0.096693625
-0.033901552
-
0.00561623
9
-
0.00284
0.01103
3
-0.015025246
-0.056887374
0
0.00686002
1
-
0.01722
-
0.00618
-0.043053301
-0.130053128
0
0.00186277
6
-
0.03616
-
0.04235
-0.019727129
-0.105360516
0.098440073
0.00185931
3
-
0.00998
-
0.05232
0.028386752
-0.057158414
-0.064538521
0.00555728
5
0.02325
6
-
0.02907
0.030248032
0.093526058
-0.143100844
-
0.00493828
2
0.01011
4
-
0.01895
0.031316744
0.023530497
-0.080042708
0.00739830
7
0.01673
7
-
0.00221
0.002411916
0.017291497
-0.042559614
-
0.00430902
5
-
0.00509
-0.0073
-0.019178329
-0.0588405
0
0.00369458
5
-
0.01997
-
0.02727
0.029234622
0.041549003
0.042559614
0.00856798
4
0.01902
-
0.00825
0.044089377
0.104778951
-0.042559614
0.00546947
6
0.02428
4
0.01603
1
0.017141065
0.035993603
-0.044451763
-
0.00425144
9
0.00768
3
0.02371
4
0.015918178
-0.010152371
-0.046520016
0.00364520
5
0.00787
1
0.03158
5
0.005585487
-0.107630664
-0.048790164
0.00181763
2
0.00767
5
0.03926
0.037583454
0.092206194
0
-
0.00363857
3
0.02476
2
0.06402
2
-0.025352594
0.17538912
-0.051293294
0.01087624
-
0.05461
0.00941
0.026291527
0.114880276
-0.30538165
0.00300030
2
-
0.00767
0.00174
4
0.004997378
0.060168521
-0.15415068
0.00537796
3
-
0.01638
-
0.01464
0.013091602
0.025226563
-0.087011377
0
0.00078
8
-
0.01385
0.022526938
-0.070010166
0.087011377
0.00594178
8
0.02664
2
0.01279
1
0.026562929
0.037457563
0
0.01002071
5
0.01391
4
0.02670
5
0.004657785
0.064078857
0.15415068
0.00058633
8
0.00125
9
0.02796
3
0.009328566
-0.013889112
-0.15415068
-
0.00646870
8
-
0.00028
0.02768
6
-0.006457531
-0.05387299
0.080042708
0.00412128
9
-
0.00361
0.02408
0.022525696
0.003683245
-0.080042708
0.00468934
0.01092
5
0.03500
5
0.009370055
-0.003683245
0
0.00408521
1
0.00301
6
0.03802
1
0.012833848
-0.056941376
-0.087011377
0.00812069
4
0.00550
1
0.04352
2
0.023847996
0.015504187
0
0.00346021
1
0.01585
0.05937
1
0.007254082
-0.023347364
0.167054085
0.00688470
9
0.01072
7
0.07009
8
-0.013200143
-0.048396541
0
0.00853976
2
-
0.01686
0.05323
6
0.008430986
0.044451763
-0.080042708
-
0.00340715
8
-
0.00409
0.04914
8
-0.043065757
-0.09531018
0
0.00340715
8
-
0.04017
0.00897
6
0.041817447
0.012959145
0.080042708
0.00565292
6
0.03752
9
0.04650
5
-0.008679239
-0.052875752
0.143100844
-
0.00225733
7
-
0.00011
0.04639
-0.012880685
-0.161720739
0.064538521
0
0.00128
8
0.04767
9
0.026587311
0.051825068
-0.064538521
0.00899893
6
0.00945
5
0.05713
3
-0.005090934
0.029852963
-0.068992871
-
0.00392707
4
-
0.01539
0.04174
7
0.015479285
-0.050261835
-0.074107972
0.00672271
4
0.00871
5
0.05046
2
0.005358555
0.125769387
0.143100844
0.00445683
2
-
0.00616
0.04430
3
-0.011982057
0.070204259
0.182321557
0.00609589
-
0.01657
0.02773
5
-0.014346117
-0.017094433
0.054067221
-
0.00332042
4
-
0.01366
0.01407
1
-0.030488015
-0.066840018
0.313657559
0.00442478
6
-
0.01395
0.00012
5
-0.05647217
0
0.037740328
0.00769657
5
-
0.06254
-
0.06241
0.03661003
-0.04717856
-0.076961041
-
0.00439078
6
0.03362
1
-
0.02879
0.009882657
0.087816206
0.182321557
0.0027465
0.00484
4
-
0.02395
-0.021810878
-0.008888947
0.587786665
0.00164428
6
0.00481
2
-
0.01913
-0.068719415
-0.0693118
0.054067221
0.00109469
1
-0.0646
-
0.08373
-0.010897043
-0.160550702
-0.054067221
-
0.00438597
2
-
0.00154
-
0.08527
0.065296043
0.059963465
0.018349139
0.00875279
1
0.05190
6
-
0.03336
2.77E-02
-0.043249984
0
0.00434783
3
0.02509
5
-
0.00827
-9.79E-06
0
0.035718083
-
0.00108518
7
-
0.00284
-
0.01111
0.007079705
-0.098630603
-0.035718083
0.00379301
5
0.00823
6
-
0.00287
0.025057856
-0.089231134
0.1198012
0.00431733
0.03343
7
0.03056
6
0.010433441
0.039220713
0.163325056
0.00590605
7
0.00794
0.03850
7
-0.008126484
0.043894194
0.027028672
0.00533904
1
-
0.01818
0.02033
2
-0.011825485
0.076733794
-0.040821995
-
0.00320000
3
-
0.02549
-
0.00515
0.008067532
0.19549202
-0.013986242
0.01009841
2
-
0.02087
-
0.02602
0.039174902
0.151476881
0.203228242
0.00316790
1
0.02885
2
0.00283
1
0.011298347
-0.024391453
0.033901552
-
0.00158269
6
0.01095
3
0.01378
4
0.018405504
-0.004123717
-0.033901552
0.00526594
2
0.00982
7
0.02361
1
0.007668838
0.02449102
0.11905936
0.00314630
6
0.00529
6
0.02890
7
-0.004869574
-0.075349437
0.04976151
-
0.00314630
6
0.00123
0.03013
7
0.010327144
0
0.019231362
0.01044941
6
0.00222
1
0.03235
9
0.014092483
-0.049007579
0.099629841
0.00414938
4
0.01750
7
0.04986
5
0.011384191
0.057665642
0.050430854
0.01998529
2
-
0.00433
0.04553
1
α= 0.005160
β1= 0.098109
β2= -0.053654
β3=0.380650
Put in AR= RSHARE_PRICES- α - β1(RLONG_TERM)- β2(RSHORT_TERM)- β3(RGDP)
.

Place new order. It's free, fast and safe

-+
550 words
Total price: $ 15.00

Our customers say

Customer Avatar
Jeff Curtis
USA, Student

"I'm fully satisfied with the essay I've just received. When I read it, I felt like it was exactly what I wanted to say, but couldn’t find the necessary words. Thank you!"

Customer Avatar
Ian McGregor
UK, Student

"I don’t know what I would do without your assistance! With your help, I met my deadline just in time and the work was very professional. I will be back in several days with another assignment!"

Customer Avatar
Shannon Williams
Canada, Student

"It was the perfect experience! I enjoyed working with my writer, he delivered my work on time and followed all the guidelines about the referencing and contents."

  • 5-paragraph Essay
  • Admission Essay
  • Annotated Bibliography
  • Argumentative Essay
  • Article Review
  • Assignment
  • Biography
  • Book/Movie Review
  • Business Plan
  • Case Study
  • Cause and Effect Essay
  • Classification Essay
  • Comparison Essay
  • Coursework
  • Creative Writing
  • Critical Thinking/Review
  • Deductive Essay
  • Definition Essay
  • Essay (Any Type)
  • Exploratory Essay
  • Expository Essay
  • Informal Essay
  • Literature Essay
  • Multiple Choice Question
  • Narrative Essay
  • Personal Essay
  • Persuasive Essay
  • Powerpoint Presentation
  • Reflective Writing
  • Research Essay
  • Response Essay
  • Scholarship Essay
  • Term Paper
We use cookies to provide you with the best possible experience. By using this website you are accepting the use of cookies mentioned in our Privacy Policy.