REGRESSION MODEL

ECONOMETRICS PROJECT
MAJOR CATALYSTS EFFECTING STOCK MARKET RETURNS

- Mudit Garg [500081851]

TABLE OF CONTENTS
1.1. INTRODUCTION .......................................................................................................................... 2
1.2. LITERATURE REVIEW................................................................................................................... 2
1.3. DESCRIPTION OF DATA ............................................................................................................... 3
Dependent Variable ............................................................................................................................ 3
Independent Variable ......................................................................................................................... 3
1.4. ECONOMETRICS METHODOLOGY .............................................................................................. 5
ORDINARY LEAST SQUARE METHOD (OLS) ........................................................................................ 5
MULTIVARIATE REGRESSION EQUATION ........................................................................................... 5
1.5. REGRESSION OUTPUT ................................................................................................................. 5
ANALYSIS ............................................................................................................................................. 6
1.6. TEST FOR MULTICOLLINEARITY .................................................................................................. 7
1.7. TEST FOR NORMALITY OF RESIDUALS........................................................................................ 8
1.8. TEST FOR HETEROSKEDASTICITY ................................................................................................ 8
1.9. TEST FOR AUTOCORRELATION ................................................................................................... 8
1.10. CONCLUSION ........................................................................................................................... 9
1.11. REFERENCES ........................................................................................................................... 9
1.1. INTRODUCTION
Stock markets are generally considered as one of the best proxy indicators of the health of an
economy. Any upside or downside in the economy affects the stock markets. Sluggishness in India’s
economy affects Sensex and Nifty. Along with slowness in economy, there are various other economic
factors as well, which bring about a significant change in the valuation of stocks. Changes in interest
rates, inflation or deflation and happenings in global markets are counted as the significant ones
among them factors.

S&P BSE Sensex is one of the most pertinent economic reflectors of our country. The stock market is
like a crystal through which the rest of the world judges the health of a country’s economy. Factors
like low labor cost and skillful manpower sectors like textile, garments, manufacturing, banking, and
insurance has made a significant contribution in the growth of the Indian economy. However, years of
stock market observations show that Nifty is affected by many determinants, whether they are
momentous occurrences in our financial vitals or the performance of international markets or simply
an occurrence of great magnitude.

The aim of this empirical research is to understand whether the movement of stock prices of the
Bombay Stock Exchange is subject to some macroeconomic variables change, in this research we
consider six macroeconomic variables: Consumer Price Index (CPI) as proxy for inflation rate, Exchange
Rate (ER), Money Supply (MS), Interest Rate (IR), Industrial Production (IIP) and Crude Oil price (COP).
The objective of this paper is to investigate the impact of macroeconomic variables on the stock
market prices of the Bombay stock exchange during the period 1994-2018. This paper is a complement
to the existing literature.

1.2. LITERATURE REVIEW

Many authors have tried to show reliable associations between macroeconomic variables and stock
returns. They identified several key macroeconomic variables which influenced stock market returns.

The aggregate performance of capital market can be easily seen by its indices that represent the
movement of stock prices being traded in capital market. (Mohi-u-Din & Mubasher, 2013) examined
stock market returns in Indian Stock Market with respect to the macroeconomic fluctuations. It
considered six variables (inflation, exchange rate, Industrial production, Money Supply, Gold price,
interest rate). The result indicates a significant relationship. (Koh, 2000) tested a similar regression
between the Singapore market and macroeconomic indicators and found a positive relationship in
money supply but negative with changes in price levels, short- and long-term interest rates and
exchange rates.

(Mahmoud Ramadan Barakat, 2016) also tried to shed light on the same relationship in two emerging
economies (Egypt and Tunisia). They resulted a causal relationship in Egypt between market index and
CPI, exchange rate, money supply, and interest rate. In, Tunisia except for CPI, which had no causal
relationship with the market index a similar result was obtained. Their research also revealed that the
four macroeconomic were co-integrated with the stock market in both countries.

The effect of some macroeconomic variables could vary from one market to another and from one
period to another. This urged a more in-depth investigation of the nature of the relationship between
macroeconomic variables and stock markets in different economies, especially after major changes.
Any changes that affect the economic conditions in general, such as removing restrictions in foreign
investment, will cause changes in the stock market and its relation to the economic factors.
Reinvestigating the nature of the relationship between the stock market and the macroeconomic

ECONOMETRICS LAB | 2
factors will be appropriate and essential for investors because the dynamics of this relationship have
changed (P, 2012).

(Sezgin Acikalin, 2008) investigated the relationships between returns in Istanbul Stock Exchange (ISE)
and macroeconomic variables of Turkish economy. They found long-term stable relationships
between ISE and four macroeconomic variables, GDP, exchange rate, interest rate, and current
account balance. A unidirectional relationship between macro indicators and ISE index. That is,
consistent with the existing literature, changes in GDP, foreign exchange rate and current account
balance influence ISE index. However, on the contrary to expectations, changes in the stock market
index do affect interest rates.

(BS, 2015) In the context of Indian Stock Market, they examined Broad Money, Call Money Rate, Crude
Oil Price, Exchange Rate, Foreign Exchange Reserve, Foreign Institutional Investors, Gross Fiscal
Deficit, Index of Industrial Production, Inflation Rate and Trade Balance and one stock market index
i.e., BSE 500 have been used to attain the objectives of the research. The two macroeconomic
variables Foreign Institutional Investors and Exchange Rate were found significant. These variables
have no relationship with closing prices of BSE 500 manufacturing firms. The study also revealed that
the Indian Stock Market was a weak form efficient because no relationship was found amongst the
variables during the study period.

1.3. DESCRIPTION OF DATA

The data set consist of one dependent variable (BSE Sensex) and 6 independent variables, the data is
monthly based taken for a period from April 1994 to September 2018.
Dependent Variable
• S&P BSE Sensex: The benchmark stock market price index of India, S&P BSE Sensex is a basket of
30 constituent stocks representing a sample of large, liquid, and representative companies.
Source: Bombay Stock Exchange (BSE)
Independent Variable
• Consumer Price Index (CPI)
Consumer price index is used as a proxy for inflation. The relationship between inflation and stock
returns can be positive or negative depending on whether the economy is facing unexpected or
expected inflation. Expected inflation happens when demand exceeds supply, causing an increase
in prices to stimulate more supply. Since this is expected by the firms, increase in prices would
also increase their earnings which would lead to them paying more dividends and hence increase
the price of their stocks as well. On the other hand, when inflation is unexpected, an increase in
price will lead to the increase in cost of living and this will shift resources from investment to
consumption. This research is based on the relationship between the unexpected changes in
economy and stock returns, thus inflation is expected to be negatively associated to stock prices.
Data Source: International Monetary Fund (IMF)

• MS- Money Supply (In Rupees Billion)

Through empirical studies it was found that there is a positive relationship between money supply
and stock prices. The argument here is that if there is an increase in money supply; it means that
money demand is increasing which is a signal of an increase in economic activity. This increase in
economic activity implies higher cash flows, which causes stock prices to rise. The money supply
is expected to be positively associated to stock returns.
Data Source: Reserve Bank India (RBI)

ECONOMETRICS LAB | 3
• Interest Rate (IR)
The money market rate is considered as a proxy for interest rate. The money market is a segment
of the financial market in which financial instruments with high liquidity and very short maturities
are traded. The money market is used by participants as a means of borrowing and lending in the
short term, from several days to just under a year. An increase in the interest rate will result in
falling stock prices since high interest rate will increase the opportunity cost of holding money,
causing substitution of stocks for interest bearing securities. The interest rate is expected to be
negatively associated to stock returns.
Data Source: International Monetary Fund (IMF)

• ER-Exchange Rate-Special Drawing Rate

An increase in exchange rate (depreciation) will cause a decline in stock prices because heavy
importer companies will suffer from higher costs due to a weaker domestic currency and will have
lower earnings, and lower share prices. As a result, the stock market, which is a collection of a
variety of companies, trends to react negatively to currency depreciation. The Exchange rate is
expected to be negatively associated to stock returns.
Data Source: Reserve Bank India (RBI)
• Crude Oil price
Any massive increase or decrease in price of crude oil has its impact on the condition of stock
markets throughout the world. India is one of the biggest importers of crude oil in the world; ranks
3rd behind U.S and China for import of crude oil. Fall in crude oil prices, therefore, helps the country
save on import bill, thereby narrowing its current account deficit. The price of crude oil is expected
to be positively associated to stock returns.
Data Source: Multi Commodity Exchange (MCX), Investing.com

• Industrial Production (IP)

A relatively average demand of stock market instruments by investors increases the amount of
fund available for industrial production. High industrial production raises corporate sales and
profits, which directly results in upsurge in trading activities in the stock market reflected by the
market index. The price of crude oil is expected to be positively associated to stock returns.
Data Source: Ministry of Statistics and Program Implementation-MOSPI

ECONOMETRICS LAB | 4
Table 1: Data Summary
Variables Explanation Data Source Expected Sign
Dependent Variable
SENSEX S&P Bombay Stock Bombay Stock Exchange NA
Exchange SENSEX (SMPI) (BSE)
Independent Variable
CPI Consumer Price Index International Monetary (-)
(CPI) as the measure of Fund (IMF)
inflation.
MS Money Supply (In Rupees Reserve Bank India (RBI) (+)
Billion)
IR Interest rate-Money International Monetary (-)
Market Rate Fund (IMF)
ER Exchange Rate-Special Reserve Bank India (RBI) (-)
Drawing Rate
COP crude oil Multi Commodity (+)
prices(international) Spot Exchange (MCX),
Price (Dollars per Barrel) Investing.com
IP Industrial Production Ministry of Statistics (+)
Index and Programme
Implementation-MOSPI

1.4. ECONOMETRICS METHODOLOGY

ORDINARY LEAST SQUARE METHOD (OLS)
The Ordinary Least Squared (OLS) econometric model is used to test the relationship
between the macroeconomic variables and the stock price index. OLS regression is a
statistical method of analysis that estimates the relationship between one or more
independent variables and a dependent variable; the method estimates the relationship by
minimizing the sum of the squares in the difference between the observed and predicted
values of the dependent variable configured as a straight line.

MULTIVARIATE REGRESSION EQUATION

𝑆𝐸𝑁𝑆𝐸𝑋𝑡 = 𝛽1 + 𝛽2 (𝐶𝑃𝐼𝑡 ) + 𝛽3 (𝑀𝑆𝑡 ) + 𝛽4 (𝐼𝑅𝑡 ) + 𝛽5 (𝐸𝑋𝑅𝑡 ) + 𝛽6 (𝐶𝑂𝑃𝑡 ) + 𝛽7 (𝐼𝐼𝑃𝑡 ) + 𝑈𝑡

1.5. REGRESSION OUTPUT

The OLS estimation equation in this regression model is,

𝑆𝐸𝑁𝑆𝐸𝑋𝑡 = 𝛽1 + 𝛽2 (𝐶𝑃𝐼𝑡 ) + 𝛽3 (𝑀𝑆𝑡 ) + 𝛽4 (𝐼𝑅𝑡 ) + 𝛽5 (𝐸𝑋𝑅𝑡 ) + 𝛽6 (𝐶𝑂𝑃𝑡 ) + 𝛽7 (𝐼𝐼𝑃𝑡 ) + 𝑈𝑡

ECONOMETRICS LAB | 5
Table 2: Regression Output [EViews]
Dependent Variable: SENSEX
Method: Least Squares
Sample: 1994M04 2018M09
Included Observations: 294

Variables Coefficient Std. Error t-Statistics Prob

C 4333.410 1388.006 3.122041 0.0020
CPI -55.7380 195.7627 -0.284722 0.7761
MS 1.3611 0.049754 27.35645 0.0000
IR -44.83383 65.52505 -0.684224 0.4944
EXR -34.70612 13.10611 -2.64886 0.0085
COP 35.21572 8.692673 4.051196 0.0001
IIP -20.65534 7.096631 -2.910584 0.0039

R – Squared 0.917999 Mean dependent var 12819.89

Adjusted R-Squared 0.916285 S.D. dependent var 9787.326
S.E. of regression 2831.819 Akaike info criterion 18.75875
Sum squared resid 2.30E+09 Schwarz criterion 18.84645
Log Likelihood -2750.536 Hannan- Quinn criter. 18.79387
F – Statistics 535.4954 Durbin-Watson stat 0.175789
Prob (F-Statistics) 0.00000

ANALYSIS
Table 1 presents the output of the Ordinary Least Square (OLS) method to show the impact of the
macroeconomics variables on stock market prices. In this case of this study shows,

• The 0.91799 value of the R-Squared implied that all the six explanatory variables together are
bale to predict 91.8% of the variable Sensex in the model.
• The F Test assumes the null hypothesis that all the variables together are insignificant. The
Prob(F-Statistics) is less than 1% and hence we reject the null hypothesis. That is, all the
explanatory variables together are significant.
• 𝛽1 = 4333.41, shows the values of Sensex at a time t when all other variables are zero. Hence
this is the intercept of the model.
• 𝛽2 = −55.738, is the slope coefficient of the variable CPI. This indicates that an increase in
CPI by one unit will cause a decrease in the Sensex by 55.738 units. The relation compliments
the theory. The t statistics assumes the null hypothesis that the variable is insignificant. The
high p value (0.776) is greater than 0.05 or 5%. Thus, CPI is an insignificant variable in the
model.
• 𝛽3 = 1.36, is the slope of the money supply. This indicates, an increase in money supply by
one unit will cause an increase in Sensex by 1.36 units. The P- value is 0 and hence the variable
is significant.
• 𝛽4 = −44.83, indicates a negative relationship between the interest rate and the returns in
stock market. The variable is insignificant induced by the high p value.
• 𝛽5 = −34.70, indicates a negative relationship between the exchange rate and the return in
the capital market. The relationship compliments the theory, and the variable is significant.
• The variable Crude oil price is positive with the stock market return and is significant in the
model. The IIP is a significant variable which shows an inverse relation to capital market
returns.

ECONOMETRICS LAB | 6
1.6. TEST FOR MULTICOLLINEARITY
Table 3: Correlation Matrix
Correlation
COP CPI EXR IIP IR MS
COP 1.0000 0.050126 0.2999392 0.614765 -0.110075 0.587726
CPI 1.0000 -0.037091 0.073144 0.033280 -0.016458
EXR 1.0000 -0.204298 0.112171 0.757754
IIP 1.0000 -0.324833 0.26501
IR 1.0000 -0.136228
MS 1.0000

Table 3 shows pair-wise correlation coefficients for each variable. No coefficient is greater than 0.8
that is, the correlation is not a serious problem in the model.

Table 4: Variance Inflation Factor

Variance Inflation Factors
Sample: 1994M04 2018M09
Included Observations: 294
Coefficient Uncentered Cantered
Variable
Variance VIF VIF
C 1926559. 70.63154 NA
CPI 38323.03 1.444542 1.010759
MS 0.002475 11.88542 4.498883
IR 4293.533 9.186899 1.196364
EXR 171.7702 20.10702 4.205420
COP 75.56256 9.933489 2.468906
IIP 50.36217 56.57205 2.755296

The values of centered VIF are all less than 10 in the table 4, this highlights that multicollinearity is not
a serious problem with the model.

ECONOMETRICS LAB | 7
1.7. TEST FOR NORMALITY OF RESIDUALS
Table 5: Jarque - Bera Test

Thus, due to high Jarque-Bera test value with low p-value we can conclude that the test for normality
is significant. Therefore, the error terms follow normal distribution as per Ordinary Least Square
assumption.

1.8. TEST FOR HETEROSKEDASTICITY

Table 6: Breusch -Pagan – Godfrey Test
Heteroskedasticity Test: Breusch -Pagan – Godfrey Test
Null Hypothesis: Homoskedasticity
F – Statistics 3.913794 Prob. F(6,287) 0.0009
Obs*R-squared 22.23612 Prob. Chi-Square(6) 0.0011
Scaled explained SS 200.6662 Prob. Chi-Square(6) 0.0000

The p value of LM stat is 0.0011, therefore there is Heteroskedasticity in the model.

1.9. TEST FOR AUTOCORRELATION

Table 7: Regression Output [EViews]

R – Squared 0.917999 Mean dependent var 12819.89

Adjusted R-Squared 0.916285 S.D. dependent var 9787.326
S.E. of regression 2831.819 Akaike info criterion 18.75875
Sum squared resid 2.30E+09 Schwarz criterion 18.84645
Log Likelihood -2750.536 Hannan- Quinn criter. 18.79387
F – Statistics 535.4954 Durbin-Watson stat 0.175789
Prob (F-Statistics) 0.00000

The Durbin Watson value in the model is 0.175. There is statistical evidence that the data is positively
autocorrelated.

ECONOMETRICS LAB | 8
1.10. CONCLUSION
The role of the stock market in the economy is to raise capital and to ensure that the funds
raised are utilized in the most profitable opportunities. This empirical report performs the
necessary analysis to answer whether changes in the identified macroeconomic variables
affect stock prices of the Bombay Stock Exchange. The research employs regression analysis
and various tests to validate these relationships. The linear regression test results show that
money supply (MS), crude oil prices (COP) and industrial production (IP) are positively and
significantly related to the stock prices of the Bombay Stock Exchange (S&P BSE SENSEX).
Besides this, there is also a negative and significant relationship between exchange rate and
stock price. This report also considers the structural break in 2008 due world financial crises
and the empirical results show that this structural break is has a significant effect on the stock
market prices. Also, the R² is 0.917999 which means that the six independent variables explain
91.8% variations in the dependent variable. It was also found that there is no multicollinearity
present between the explanatory variables. As per OLS assumption, Jarque-Bera test proved
that the residuals follow normal distribution.

Based on the above overall analysis, it can be concluded that four out of the six selected
macroeconomic variables are relatively significant and likely to influence the stock prices of
the Bombay Stock Exchange.

1.11. REFERENCES
BS, G. K. (2015). An Impact of Macroeconomic Variables on the functioning of Indian Stock Market: A
Study of Manufacturing Firms of BSE 500. Journal of Stock & Forex Trading.

Koh, R. C. (2000). A vector error correction model of the Singapore stock market. International
Review of Economics & Finance.

Mahmoud Ramadan Barakat, S. H. (2016). Impact of Macroeconomic Variables on Stock Markets:

Evidence from Emerging Markets. International Journal of Economics and Finance.

Mohi-u-Din, S., & Mubasher, H. M. (2013). Macroeconomic Variables on Stock Market Interactions:
The Indian Experience. Advances in Management.

P, P.-K. N. (2012). The impact of macroeconomic fundamentals on stock prices revisited: An evidence
from Indian data. Eurasian Journal of Business and Economics.

Sezgin Acikalin, R. A. (2008). Relationships between stock markets and macroeconomic variables: an
empirical analysis of the Istanbul Stock Exchange . Investment Management and Financial
Innovations.

ECONOMETRICS LAB | 9