Академический Документы
Профессиональный Документы
Культура Документы
on
QUANTITATIVE ANALYSIS OF LARGE STOCK
MARKET CRASHES
Submitted in partial fulfillment for the award of the degree
of
MASTER OF BUSINESS ADMINISTRATION
in
FINANCE
by
NEELANJAN MAITI
(Professor)
MARCH 2020
DECLARATION
The material borrowed from similar titles other sources and incorporated in
The matter embodied in this project report has not been submitted to any
original work and it has not been presented earlier in this manner. This
Signature Date:
own and it is an original work and not copied from any resource.
Thanking you.
Yours Truly,
Assistant Professor
lot of Research and I came to know about so many new things I am really
thankful to them.
Secondly, I would also like to thank my parents and friends who helped
I am making this project not only for marks but to also increase my
NEELANJAN MAITI
ABSTRACT
strength. Prices of stock market are volatile in nature and are affected by various
factors like inflation, economic growth, individual risk etc. Thus, the prices of a
share market are mainly driven by demand and supply. High demanding stocks
will see an increase in price whereas low demanding stocks will see a decrease
in price as they will be sold heavily. These significant drop in price if occurring
rapidly can lead to a stock market crash. occur due to some major catastrophic
which has been there for long enough to cause the damage. A crash is generally
considered when the majority of stocks decline rapidly and for a period of time.
In this report, we study and analyze the reasons behind the crashes and what were
the consequences. Among the many quantitative tools, we use regression analysis
which can be used to predict the market movement by analyzing the historical
data so that investors can anticipate the market and can make appropriate buy or
sell decisions.
TABLE OF CONTENTS
INTRODUCTION
1.1 OBJECTIVES
• Predicting market movements by using regression analysis which can help an
investor to make informed decisions
1.2 NECESSITY
• Stock market crash erodes investors’ wealth
• This project provides exposure on how to predict market movement more
accurately
• It allows the investor to make appropriate buy and sell decisions
1.3 SCOPE
• This project is carried out taking major global indices into consideration
• Regression analysis is used which is best suited for predicting markets
• A long time period is taken to ensure the accuracy of future predictions
1.4 INDIAN STOCK MARKET
There are two types of investors: those who are aware of the investment opportunities
available in India and those who are not. A stock market is a place where buyers and
sellers of stocks come together, physically or virtually. Participants in the market can
be small individuals or large fund managers who can be situated anywhere around the
globe. Investors place their orders to the professionals of a stock exchange who
executes these buying and selling orders. The stocks are listed and traded on stock
exchanges. Some exchanges are physically located, based on open outcry system
where transactions are carried out on trading floor. The other exchanges are virtual
exchanges whereas a network of computers is composed to do the transactions
electronically. The whole system is order-driven, the order placed by an investor is
automatically matched with the best limit order. This system provides more
transparency as it shows all buy and sell orders. The Indian stock market mainly
functions on two major stock exchanges, the BSE (Bombay Stock Exchange) and NSE
(National Stock Exchange). In terms of market capitalization, BSE stood at 10th
position with market capitalization of $2.05T as on November, 2018 and NSE at 11th
position with market capitalization of $2.03T as on November, 2018.
Generally speaking, rising interest rates are a negative catalyst for stocks
and therefore the economy generally. This is very true for income-
focused stocks, like land investment trusts (REITs). Investors buy these
stocks specifically for his or her dividend yields, and rising market
interest rates put downward pressure on these stocks. From an economic
standpoint, higher interest rates mean higher borrowing costs, which
tends to hamper purchasing activity, which may successively cause
stocks to dive. So, if the 30-year mortgage rate were to spike to, say, 6%,
it could dramatically hamper the housing market and cause homebuilder
stocks to require successful
• Political risks
While nobody features a ball, which will predict the longer term, it is a
safe bet that the stock exchange wouldn't love it much if the U.S. went to
war with China. Markets like stability, and wars and political risk
represent the precise opposite. For instance, the Dow Jones Industrial
Average dropped by quite 7% during the primary trading session
following the Sept. 11, 2001, terror attacks, because the uncertainty
surrounding the attacks and therefore the next moves spooked investors.
• Tax changes
The market crash effects both the economy and the individual investors on several
fronts:
• Wealth effect
The first impact is that people with shares will see a fall in their wealth.
If the fall is significant, it will affect their financial outlook. If they are
losing money on shares, they will be more hesitant to spend money; this
can contribute to a fall in consumer spending. However, this effect should
not be given too much importance. Often people who buy shares are
wealthy and prepared to lose money; their spending patterns are usually
independent of share prices, especially for short-term losses. Also, only
around 10% of households own shares – for the majority of consumers,
they will not be directly affected by a fall in share prices. The wealth
effect is more prominent in the housing market. (e.g. falling house prices
affect more consumers)
• Effect on pension
• Confidence
Falling share prices can hamper firm’s ability to raise finance on the stock
market. Firms who are expanding and wish to borrow often do so by
issuing more shares – it provides a low-cost way of borrowing more
money. However, with falling share prices it becomes much more
difficult.
• Bond market
• Mid to long term investors or buy side firms who purchase stocks in large
quantities but do not want to influence stock prices with discrete large
volume investments
• Short term traders and sell participants who benefit from automated trade
execution in addition to creating liquidity
1.10 REGRESSION
Linear Regression analysis found its way in the finance sector as a method in
modeling a single relationship by analyzing two different variables. The most
common variables used for creating this relationship are price and time. The
correlation between these two factors forming this relationship allows the modeling
of capital asset pricing and additionally applies to the concept of quantifying the risk
of investment. Regression models come in all shapes and sizes. Some of the important
models that are used in today’s trading environment are:
LITERATURE REVIEW
Léon et al (2008), studied the relationship between expected stock market returns and
volatility in the regional stock market of the West African Economic and Monetary
Union called the BRVM. The study revealed that expected stock return has a positive
but not statistically significant relationship with expected volatility and volatility is
Mubarak and Javid et al (2009), investigated the relationship between trading volume
and returns and volatility of Pakistani market. The findings suggested that there was
significance effect of the previous day trading volume on the current return and this
implied that previous day returns and volume has explanatory power in explaining the
Srinivasan and Ibrahim et al (2010), attempted to model and forecast the volatility of
the SENSEX Index returns of Indian stock market. Results showed that the symmetric
Index return rather than the asymmetric GARCH models, despite the presence of leverage
effect.
market using different GARCH models. First, the standard GARCH approach was used
to investigate whether stock return volatility changes over time and if so, whether it was
predictable. Then, the E-GARCH models were applied to investigate whether there is
asymmetric volatility. It was found that the volatility is an asymmetric function of past
Gupta et al (2013), aimed to understand the nature and different patterns of volatility in
Indian stock market on the basis of comparison of two indices which are BSE index,
SENSEX and NSE index, NIFTY. GARCH models were used to see the volatility of
Indian equity market and it was concluded that negative shocks do have greater impact on
conditional volatility compared to positive shocks of the same magnitude in both indices
i.e. SENSEX and NIFTY of the Bombay Stock Exchange and National Stock Exchange.
Gahan et al (2012), studied the volatility pattern of BSE Sensitive Index (SENSEX) and
NSE Nifty (Nifty) during the post derivative period. The various volatility models were
developed in the present study to get the approximately best estimates of volatility by
recognizing the stylized features of Stock market data like heteroscedasticity, clustering,
asymmetry autoregressive and persistence. When compared, it was found that there was
difference between the volatility of pre and post derivative period. Conditional volatility
determined under all the models for SENSEX and Nifty were found to be less in post
Singh et al (2008), studied some of the issues related to the estimation of beta. It was
found that beta varies considerably with method of computation and the major reason for
variation seems to be the interval between data points. While the correlation between
weekly and daily betas was very high, this was not the case with weekly and monthly
betas. The variability of betas was higher with longer interval periods and more stocks
Srinivasan (2010) examined the random walk hypothesis to determine the validity of
weak-form efficiency for two major stock markets in India. He suggested that the Indian
stock market do not show characteristics of random walk and was not efficient in the weak
Srinivasan et al (2010), examined the random walk hypothesis to determine the validity
of weak-form efficiency for two major stock markets in India. He suggested that the Indian
stock market do not show characteristics of random walk and was not efficient in the weak
Bhat et al (2014), focused on analyzing and comparing the efficiency of the capital
markets of India and Pakistan. The results derived by using various parametric and non-
parametric tests clearly reject the null hypothesis of the stock markets of India and
Pakistan being efficient in weak form. The study provides vital indications to investors,
hedgers, arbitragers and speculators as well as the relevance of fundamental and technical
analysis as far as the trading/investing in the capital markets of India and Pakistan is
concerned. A gap is seen between the studies as some are in favor that Indian stock market
are weak form efficient while other are against it, so this gap helped in formulating another
objective which is to seek the weak form efficiency of Indian stock market.
Kiymaz and Berument et al (2003), investigated the day of the week effect on the
volatility of major stock market indexes. It was found that the day of the week effect was
present in both return and volatility equations. The highest volatility occurred on Mondays
for Germany and Japan, on Fridays for Canada and the United States, and on Thursdays
for the United Kingdom. For most of the markets, the days with the highest volatility also
Chander and Mehta et al (2007), emphasized on that investors and analysts are unable
efficient markets. It was seen whether anomalous patterns yield abnormal return
consistently for any specific day of the week even after introduction of the compulsory
rolling settlement on Indian bourses. The findings recorded for post- rolling settlement
period were in harmony with those obtained elsewhere in the sense that Friday returns
Chia and Liew et al (2010), studied the existence of day-of-the-week effect and
asymmetrical market behavior in the Bombay Stock Exchange (BSE) over the pre- 9/11
and post-9/11 sub-periods. They found the existence of significant positive Monday effect
and negative Friday effect during the pre-9/11 sub-period. Moreover, significant day-of-
the-week effect was found present in BSE regardless of sub- periods, after controlling for
Sah et al (2010), believed the main cause of seasonal variations in time series data is the
change in climate. The study found that daily and monthly seasonality were present in
NIFTY and NIFTY Junior returns. It was found that Friday Effect in NIFTY returns while
NIFTY Junior returns were statistically significant on Friday, Monday and Wednesday.
In case of monthly analysis of returns, the study found that NIFTY returns were
Swami et al (2011), investigated four calendar anomalies, viz., Day of the Week effect,
Monthly effect, Turn of the month effect and Month of the year effect across five countries
of South Asia. The day of the week effect, was found to exist in Sri Lanka and Bangladesh;
and the intra-month return regularity, in terms of Monthly effect and Turn of the month
effect, was present in the Indian market. The anomalous behavior was not pervading
across the five countries and there was little influence of one market over the other, so far
Kaur et al (2004), investigated the nature and characteristics of stock market volatility in
Indian stock market in terms of its time varying nature, presence of certain characteristics
such as volatility clustering, day-of-the-week effect and calendar month effect and
whether there existed any spillover effect between the domestic and the US stock markets.
It showed that day-of-the-week effect or the weekend effect and the January effect were
not present.
Dasgupta et al (2014), found only one co-integration, i.e., long-run relationships and also
short-run bidirectional Granger relationships in between the Indian and Brazilian stock
markets. It was found that the Indian stock market has strong impact on Brazilian and
Russian stock markets. The interdependencies (mainly on India and China) and dynamic
linkages were also evident in the BRIC stock markets. Overall, it was found that BRIC
stock markets are the most favorable destination for global investors in the coming future
and among the BRIC the Indian stock market has the dominance. On the basis of above,
it is seen that a gap is prevalent. This gave an origin to the objective of whether Indian
stock market is interdependent on international stock markets or not so that this gap can
be filled.
and three developed (Hong Kong, Japan and Singapore) and four emerging (China, India,
Malaysia and Russia) markets of Asia. While bivariate Johansen co- integration test
from Australia to Asian markets disappears after the crisis. Results of VAR models
markets.
dependence in three major markets of South Asia, India, Sri Lanka and Pakistan. It was
realized that merely identifying non-linear dependence was not enough. The application
of the BDS test strongly rejects the null hypothesis of independent and identical
distribution of the return series as well as the linearly filtered return series for all the
RESEARCH OBJECTIVE
3.1 OBJECTIVES
• To analyze the factors responsible behind a stock market crash
• To calculate the recovery time of a market after its crash
• To discuss practical implications of using quantitative tools
• To predict the market movement to make appropriate buy and sell decisions
CHAPTER 4
RESEARCH METHODOLOGY
4.1 METHODOLOGY
4.1.2 Timeline
The various reasons for a market crash are studies thoroughly as well as the
consequences are also looked into so as to understand how the economy as
well as the various retail and institutional investor are affected
4.1.5 Prediction
The various predictions are carried out which can determine whether ca
crash will occur or not in the near future
Literature survey: Books and journals required for the project are collected
The responsible factors and the after effects of a crash are studied
RESEARCH ANALYSIS
The stock market crash of 2008 was the biggest single-day drop in history
up to that point. The aftermath of this catastrophic financial event wiped out
big chunks of Americans’ retirement savings and affected the economy long
after the stock market recovered. The financial turmoil caused by the crisis
impacted many sectors, leading to massive job losses and mortgage defaults.
As investment firms collapsed and automakers stood on the verge of
bankruptcy, the federal government stepped in and “bailed out” company
after company.
The Glass Steagall Act of 1933 allowed banks, securities firm and
other insurance companies to enter into each other’s markets
resulting in the formation of the bank that was too big to fail.
Also, the SEC in 2004 weakened the capital requirements thus
encouraging the bank to invest more money in MBS. Although
the decision took by SEC resulted in enormous profit but it also
exposed the portfolio to a significant risk.
• Rating agencies
One of the main reasons of this crisis were the rating agencies
who classified subprime securities as investment grade. Part of it
was incompetence and part of it was conflict of interest as the
rating agencies were paid by issuers to rate the securities.
The Dow Jones Industrial Average plunged by 54% from its high
to 6440 points in the first quarter of March,2009 which was its
all-time low. Both SENSEX and NIFTY crashed by 60% to 8000
points and 2500 points respectively in the first quarter of march
• Recovery
The crashes continued till the first quarter of the following year
and all the major indices took about 5-6 years to recover from
their all-time lows to their previous highs
As the prices of dot-com stocks peaked, the bubble burst and the slide
began. The consequences of this burst were:
• Dot com companies went bankrupt
• Recovery
The crashes continued for the next couple of years before all the
major indices could recover and regain to their previous highs
Out of the various available methods, regression analysis is used for this purpose
as it is one of the widely used methods to predict the market movements. This is
carried out in Excel.
For predicting the future point movement of SENSEX, the historical monthly
closing of 20 years is taken into consideration. An equal time interval is taken. The
data is exported to excel and then taking the data range of Timeline and Values,
forecast analysis was carried out. A confidence interval of 95% was considered and
the upper and lower confidence bounds were defined. The forecast was done for the
next five years and accordingly the chart and table were prepared displaying the
monthly closing prices
70000
60000
50000
40000
POINTS
30000
20000
10000
0
1-Nov-00
1-Nov-05
1-Nov-10
1-Nov-15
1-Nov-20
1-Jul-02
1-May-03
1-Jul-07
1-May-08
1-Jul-12
1-May-13
1-Jul-17
1-May-18
1-Jul-22
1-May-23
1-Jan-00
1-Sep-01
1-Mar-04
1-Jan-05
1-Sep-06
1-Mar-09
1-Jan-10
1-Sep-11
1-Mar-14
1-Jan-15
1-Sep-16
1-Mar-19
1-Jan-20
1-Sep-21
1-Mar-24
TIMELINE
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
1/1/2000
1/1/2001
1/1/2002
1/1/2003
1/1/2004
Values
1/1/2005
1/1/2006
1/1/2007
1/1/2008
1/1/2009
1/1/2010
1/1/2011
Lower Confidence Bound(Values) 1/1/2012
1/1/2013
1/1/2014
1/1/2015
1/1/2016
TIMELINE
1/1/2017
1/1/2018
1/1/2019
Fig 5.6 Predicted movement of DJIA
Forecast(Values)
1/1/2020
1/1/2021
1/1/2022
1/1/2023
1/1/2024
Upper Confidence Bound(Values)
1/1/2025
1/1/2026
1/1/2027
1/1/2028
1/1/2029
12/31/2029
CHAPTER 6
6.1 Limitations
The method of regression analysis is widely used by many to predict the future
movements. It can be both used in Excel as well as used in many programming
languages like Python and Java. Although the prediction is very accurate but still this
method has its own limitations as compared to the other models.
• It is sensitive to outliers
In this project regression analysis has been used to predict the market
movement but it can also be used for various other types of tasks.
RECOMMENDATIONS
CONCLUSION
The conclusion that can be drawn from this project is that stock market crashes occur
due to a variety of factors whether it is due to panic or whether it is due to economic
distress but if quantitative tools are properly used to predict the markets, the any
individual can predict whether the market is going to rise or fall and accordingly make
appropriate buy or sell decisions
CHAPTER 9
REFERENCES
1. Bahng, S. (2003). The Response of the Indian Stock Market to the Movement of Asia's
Emerging Markets: From Isolation Toward Integration? Global Economic Review, 32
(2), 43-58.
2. Chander, R., & Mehta, K. (2007). Anomalous Market Movements and the Rolling
Settlement. Vision: The Journal of Business Perspective, 11(4), 31-44. Cootner, P. H.
(1962). Stock prices: Random vs. Systematic Changes. Industrial Management Review,
3(2), 24–45.
3. Dasgupta, R. (2014). Integration and Dynamic Linkages of Indian Stock Market with
BRIC-An Empirical Study. Asian Economic and Financial Review, 4 (6), 715-731.
Goldsmith, R. W. (1971). Capital Markets and Economic Development. National
symposium on Development of Capital Markets. September 1971.
4. Gupta, R. (2010). Movement of SENSEX: Domestic and International Factors. Finance
India, 24 (1), 85-96.
5. Karmakar, M. (2007). Stock Market Asymmetric Volatility and Risk-Return
Relationship in the Indian Stock Market. South Asia Economic Journal, 8, 99- 116.
6. Kaur, H. (2004). Time Varying Volatility in the Indian Stock Market. Vikalpa-The
Journal of Decision Makers, 29, 25-42.
7. Khan, A. Q., Ikram, S., & Mehtab, M. (2011, June). Testing weak form market
efficiency of Indian capital market: A case of national stock exchange (NSE) and
Bombay stock exchange (BSE). African Journal of Marketing Management, 3(6), 115-
127.
8. Kiymaz, H., & Berument, H. (2001). The day of the week effect on stock market
volatility. Journal of Economics & Finance, 25(2), 181-193.
9. Kumar, R., & Dhankar, R. S. (2009). Asymmetric Volatility and Cross Correlations in
Stock Returns under Risk and Uncertainty. Vikalpa-The Journal of Decision Makers,
34, 25-36.
10. Leon, N. K. (2008). An Empirical Study of the Relation Between Stock Market Returns
and Volatility in the BRVM. International Research Journal of Finance and Economics,
14.
11. Madhavi. (2014). An Evaluating Study of Indian Stock Market Scenario with Reference
to its Growth and Inception Trend Attempted by Indian Investors: Relation with LPG.
GALAXY International Interdisciplinary Research Journal
, 2 (2), 172-179.
12. Majid, M.S.A., Meera, A.K.M., & Omar M.A. (2008). Interdependence of ASEAN-5
Stock Markets from the US and Japan. Global Economic Review: Perspectives on East
Asian Economies and Industries, 37:2, 201-225, DOI:10.1080/12265080802021201
13. Mubarik, F., & Javid, A. (2009). Relationship between Stock Return Trading Volume
and Volatility: Evidence from Pakistani Stock Market. Asia Pacific Journal of Finance
and Banking Research , 3 (3).
14. Mukherjee, K., & Mishra, R. K. (2007). International Stock Markets Integration and its
Economic Determinants: A Study of Indian and World Equity Markets. Vikalpa-The
Journal of Decision Makers, 32, 29-44.
15. Ranganatham, M., & Subramanian, V. (1993). Weak Form of Efficient Markets
Hypothesis: A Spectral Analytical Investigation. Vikalpa, 18 (2).
16. Singh, S., & Sharma, G. D. (2012). Inter- Linkages between Stock Exchanges: A study
of BRIC Nations. International Journal of Marketing, Financial Services, Management
Research, 1 (3). Online copy available at: indianresearchjournals.com.
17. Srinivasan, P., & Ibrahim, P. (2010). Forecasting Stock Market Volatility of BSE-30
Index Using GARCH Models. Asia-Pacific Business Review, 6 (3), 47-60.
CHAPTER 10
ABSTRACT