Impacts of Macroeconomic Variables on the Stock Market Returns of South Asian Region

This paper is intended to find out whether macroeconomic variables may impact on stock market as well as whether such impact has any country specific pattern. Stock market return was taken as dependent variable and real interest rate, inflation rate, GDP growth rate, foreign currency reserve growth rate, fiscal deficit, FDI to GDP ratio, exchange rate were taken as independent variables. Data-set was covered from 1993 to 2019 for five South Asian countries which were Bangladesh, India, Pakistan, Sri Lanka, and Nepal. Pattern of stock market as well as macro conditions of these countries was observed and it was found that some relationships exist between the stock market returns and these chosen independent variables. Unit root test, Heteroscedasticty test, autocorrelation test, Hausman test is conducted to authenticate and clarified data to investigate relationship nature. Granger Casualty test indicated that there exist cause and effect relationship between GDP growth rate, exchange rate, and stock market returns. Finally, regression test reveals that inflation rate and foreign currency reserve growth rate have significant impact on the stock market returns. It was expected to have unique nature of different countries having versatile impact on dependent, so additionally fixed effects model and random effects model were run and it was found that random effects model is statistically appropriate through conducting Hausman test. The test reveals that GDP growth rate, foreign currency reserve growth rate, and fiscal deficit positively impact on the stock market returns and these also support the literature review. Interest rates, inflation rate, FDI to GDP ratio and exchange rate have negatively impact the stock market return where only interest rate, inflation rate & exchange rate.

A lot of interest has been generated in the field of stock market. The performance of the stock market varies due to many reasons. This paper focuses on understanding the relationship between major macroeconomic variables and the stock market returns with the passage of time. Explorative studies have been measured through gathering the required information. Different required variables have been taken for the analysis. The relationship between change in the performance of these stock markets and variables in the timeframe are examined.

Objectives of this Study
1. To determine the impact of major macroeconomic variables of 5 South Asian countries. 2. To analyze the influences of the variables on their stock market returns.
3. To understand the determinants of stock market returns from these variables.

Rationale of this Study
South Asian region has made significant stride towards the development. It can be big investment fields for the foreign investors. But the performance of the stock market returns doesn't remain the same over time. Due to change in the major macroeconomic variables, the overall performance of the stock markets is influenced. This study has provided the idea about the influence of major macroeconomic variables towards the stock market of South Asian regions.

Scope of this Study
The study of this paper provides the understanding to learn more about the impacts of macro variables on the stock market returns of five south Asian countries for the last 25 years. It also provides the opportunity to know details about stock market condition of these five countries. Different techniques and regression analysis have been come through lessons with the literature review of impacts of macro variables on stock market returns of these countries over time.

Limitations of the Study
Despite the limitations of my knowledge, thinking power and skills, I tried to give best effort to understand the macro variables impacts. I had to make some assumptions because of unavailability of information. There was some confusion regarding some data and calculation process where I used my own opinions and judgments to take decision. Indices of NEPSE for 1993 & 1994 were assumed.

Literature Review
Lots of researches have been conducted to understand the impacts of major macroeconomic variables. Sharma & Mehendru (2010) found that inflation rate and exchange rate negatively impact the stock price index where foreign exchange reserve and gold price positively impact the market. Impacts of exchange rate and gold price were found significant. Ali

METHODOLOGY:
3.1 Techniques of Analysis -For analyzing and examining the impacts of macro variables on the stock market returns, regression analysis which is a great statistical tool has been used. For making the structure more specific and easier access, different independent and dependent variables have been taken.

Regression
Analysis & Model -Conditional expectation of dependable variable given independent variables is estimated through regression analysis. For finding out the regression analysis, different techniques have been developed. Linear regression and ordinary least squares regression are much familiar and parametric. Different factors of following formulas have been used. These models are used when these are less expensive in terms of money or less time consuming in terms of decision. Some examples of the usage of regression models will be presented before describing the details of modeling process. The information going to be used for making the prediction and the information going to be predicted have to be obtained from the sample of objects or the individuals. Through a linear transformation, the relationship between two parts of information is modeled.
Where, Y is the dependent variable, is the intercept term, are the n coefficients for independent variables, Ɛ is the error term. For the impact of macro-variables analysis of five south Asian countries over the last 25 years, Y= Stock Market Returns (Stock market index returns of five south Asian countries).

Sources of Data & Information -
The data and required information of the impacts of macro variables on stock market returns have been collected from different websites. Data are of last 25 years. Many websites have been browsed for gathering the information. Some data have been collected from government websites.

Variables
Dependent Variable: The dependent variable for the analysis is stock market return that has been calculated through the yearly indices from 1993 to 2019. For stock market returns of Bangladesh, general index of DSE from 1993 to 2012 and DSEX index from 2013 to 2019 have been used.
For India, BSE Sensex index have been used. For Pakistan, KSE100 index have been used. For Sri Lanka, all share price index of CSE has been used. Returns have been calculated by using the formula ln [(y1-yo)/yo].

Independent Variables:
Real interest rate, inflation rate, GDP growth rate, foreign currency reserve growth rate, fiscal deficit, FDI to GDP ratio, exchange rate are the independent variables. Data consist of the years from 1993 to 2019.

Stock Markets
South Asia is the southern part of Asia in the world. South Asian countries cover almost 1891 million people that make the most densely populated and most populous geographical region in the world. Total Nominal GDP size is $3.12 trillion.

Opportunities, Challenges of Stock Markets of South Asian Region
South Asian countries have some potential of better growth aspects in the upcoming future. This region is one of the fastest growing regions in the whole world. Poverty rates are on the decrease although this region covers almost 40% of world's total poverty. Earlier illiteracy was very strongly prevalent in this region. South Asian region is least economically integrated area in the world. Overall financial markets are getting improved. Capital inflows are received from outside parts. Greater forms of innovation in financial markets are coming. Cheaper forms of financing are proliferating of the market growths.
Transactions costs are decreasing bring to higher returns. Bilateral issues are somewhat lopsided in this region. Higher poverty dispels the aspect of development (Apriyanti, 2020). Regional policies are not strong enough to cooperate themselves. Thus it's hindering the expected regional development. Lack of information and also the interest has limited the opportunities and growths in this region. Strategic weaknesses have weakened the overall market efficiency.
There are seen lacks of interest in market developments and also the innovations. Low awareness in private sector initiatives has deteriorated the situation. Market should be more accessible for the foreign investors. Exchange control should be more efficient. Trading should be more transparent. Regional cooperation should be increased.

ANALYSIS AND RESULTS:
5.1 Descriptive Statistics -For describing the basic features of data in this analysis, descriptive statistics have been done. Total number of observations is 125 for every variable are given below ( Table 1).

Unit Root
Testing -When a time series faces unit root problem, the systematic pattern becomes unpredictable. Because of panel data, Levin Lin Chu unit root test method has been used to testify whether any unit root problem exist in the chosen variables. The findings from the test are given below ( Table 2).

Granger Causality
Test -By using empirical data sets for finding patterns of correlation, this method follows probabilistic account of the causality. For identifying the lagging, leading and the coincidence of macroeconomic variables for testing the stock market performance, Granger causality test enables to realize this.  GDP growth rate and exchange rate have a predictive relationship at 10% level of significance. The overall Granger causality tests indicates no significant relationship between stock market returns and the rest macroeconomic variables in strongly indicates the evidence of the informational inefficiency in the market.

Correlation Matrix & VIF Test
When the correlation is 1.00, it indicates two variables are perfectly correlated. All these independent variables are moderately correlated with one another. Variance Inflation Factor (VIF) indicates the ratio of the variance in any model with the multiple terms divided by variance of a model with one term alone. The Variance Inflation factor (VIF) test in Stata software shows that the mean is 1.29. It assures that there is no multicollinearity problem.

Autocorrelation or Serial Correlation -
Through conducting Breusch Godfrey test for autocorrelation in Stata, p value was found as 0.5662 that clearly tells no serial correlation is exist in the model.

Regular Multiple Regression, Fixed Effects
Model, Randome Effects Model -The R squared value becomes 0.0947. The adjusted R squared becomes 0.0405 and the root MSE becomes 0.29292. Total number of observations was 125. It indicates that among all the independent variables, the impact of foreign currency growth rate becomes significant at 5% level of significant. And the impact of inflation rate becomes significant at 10% level of significance. It's found that, some countries have the highly different values while in some area countries are best for their different other variables.
In fixed effects model, Here R square is within 0.0996, between 0.5081, overall 0.0903. F(7,113)= 1.79. Prob>F= 0.096. Corr = -0.1603. It means all the independent varaibles can impact by 9.96% on the stock market return. P value 0.09 indicates the model is not weak. Random effects model follow GLS regression equation. Generalized Least Square regression method is being followed by random effect model because of cross sectional data.
Here, R square is within 0.0975, between 0.5465, overall 0.0947. Wald chi2 (7)=12.23. Prob>chi2= 0.0932. Corr=0. It means all the independent varaibles can impact by 9.75% on the stock market returns. P valoe 0.09 indicates the model is not weak. The comparison among the multiple regression model, fixed effects model and random effects model is given below ( Table 3).
From the comparison, we can understand that random effects model and multiple linear regressions provide almost the same result. The coefficients of the independent variables under these two methods become almost same. The p values under these models become similar too.

Impacts of Different Variables on Market Return
Real Interest Rate: The coefficient of real interest rate is -0.7779. It indicates that there exists negative relationship between the stock market return and the real interest rate. Due to 1% increase in real interest rate, it leads to 0.77% decrease on the stock market return. P value 0.292 which is higher than 0.05 that indicates the impact isn't significant.
Inflation Rate: The coefficient of real inflation is -1.3787. It indicates that there exists negative relationship between the stock market return and the inflation rate. Due to 1% increase in inflation rate, it leads to 1.38% decrease on the stock market return. P value 0.124 which is higher than 0.05 that indicates the impact isn't significant. But at 10% level of significance, we may consider it as significant.

GDP Growth Rate:
The coefficient of GDP growth rate is 1.7341. It indicates that there exists positive relationship between the stock market return and the GDP growth rate. Due to 1% increase in GDP growth rate, it leads to 1.73% increase on the stock market return. P value 0.192 which is higher than 0.05 that indicates the impact isn't significant.

Foreign Currency Reserve Growth Rate:
The coefficient of foreign currency reserve growth rate is 0.1646. It indicates that there exists positive relationship between the stock market return and the foreign currency reserve growth rate. Due to 1% increase in foreign currency reserve growth rate, it leads to 0.1646% increase on the stock market return. P value 0.021 which is lower than 0.05 that indicates the impact is significant. For collecting the coefficient, probability factor is used for the regression analysis tools and software. From the analysis, it's found that some independent variables have positive impact on the market return and the degree of influences is significant. In comparison with that, some independent variables have negative effect on market return and the degree of influence is insignificant. The countries are much competitive in market but their performance is somewhat not equal to other markets in many other countries. A little bit significant impact is noticed on stock market returns for different variables.

FINDINGS OF THE STUDY:
The purpose of the paper was to analyze the impacts of macroeconomic variables on the stock market returns of selected South Asian countries. All the information is gathered from different sources. All data are categorized with different variables. The literature review provides the understanding of the impacts of major macroeconomic variables on the stock market returns based on previous researches. The first objective of the paper was to determine the impact of major macroeconomic variables of the chosen countries on the stock market returns. It's found that inflation rate and foreign currency reserve growth rate have significant impacts on the stock market returns. Second objective was to analyze the influences of the chosen independent variables on the stock market returns. By conducting unit root testing in Stata, no problem was found with the chosen independent variables.
Levin Lin Chu unit root test was applied in it. All the variables were found as stationary. Third objective of the paper was to identify the major determinants of stock market returns. It's found that exchange rate has the most influence among all the independent variables on the stock market returns due a small change in it. As inflation rate and foreign currency reserve growth rate impact significantly on the stock market returns, these are strong determinants of the stock market performance.