Relationship between Macroeconomic Variables and Stock Market: Case Study from Malaysia

This study investigate the relationship between macroeconomic variables and FTSE Bursa Malaysia KLCI, the samples are divided into 2 groups such as foreign macroeconomic variables and local macroeconomic variables, foreign macroeconomic variables consist of Gold Bullion LBM price and Dow Jones Index, meanwhile local macroeconomic variables consist of Consumer Price Index, Base Lending Rate, Exchange Rate. This study employs data from Jan 2000 to Dec 2013 which contains a monthly data set of 168 observations. There are 3 methodologies used in this study to investigate the relationship, the first test is Unit Root test which used to test the stationary of each variable, the results indicate that all the variables are stationary in first difference, this is important to use stationary variables because if the variables are not stationary, it might lead to spurious regression. The second methodology is Johansen & Juselius Co-integration test to investigate the long run relationship among these variables, the results show that the foreign macroeconomic variables and local macroeconomic variables have long run relationship with KLCI and significant. Next, this study will investigate the short run relationship between macroeconomic variables and KLCI, the results indicate that Gold, BLR and CPI can granger cause KLCI and significant at 1%, 5% significance level respectively.


Introduction
The stock market is one of the sources that play an important role in contributing to the economic growth of a country. Government, industry, central bank and investor of the country are vital to keep a close watch on the happenings of the stock market. According to Zukarnain and Sofian (2012), the stock market should be theoretically closely linked with the macroeconomic variables of the country because stock prices are the discounted present value of expected future cash flows. Therefore, there were a few numbers of studies that have been conducted to determine the existence of the relationship between macroeconomic variable and stock market. There are many important macroeconomic variables to represent the economic growth of a country such as GDP, CPI, interest rate, money supply and others. They are similar with the stock market where both also are important to the country growth. Therefore it is worthy for us to determine whether the macroeconomic variable can explain the stock market or vice versa.
The knowledge of the factors which can influence the behavior of stock market and macroeconomic variables has attracted the attention not only from investors but also to the policy makers for a long time but it is still hardly to determine whether which macroeconomic variables can influence the stock market directly. For investors such as retail investors or institutional investors, by knowing the relationship between the macroeconomic variables and stock market can help them appropriately forecast stock price movement, whereas for government sector such as policy makers, it is important for them to identify the relationship between stock market and macroeconomic variables because they can use the stock market as a leading indicator to predict future macroeconomic variables if the stock market leads macroeconomic variables.
This study not only include with the local macroeconomic variables but also with foreign macroeconomic variables in order to study the effect of it towards Malaysia stock market. The local macroeconomic variables adopted in this study are as CPI, ER and BLR, whereas for foreign macroeconomic variables adopted in this study are Dow Jones Industrial Average (DJIA) and Gold.

Problem Statement
Recently, there are many studies to investigate the relationship between stock market and the country macroeconomic variables. The main reason is because the stock market has been recognized to have a prominent role in a country economic performance. It is very crucial for investors in the equity market, bank in the financial institution as well as policy makers to discover the relationship between macroeconomic variables and the movement of the stock market because it can help investors and financial institutors to forecast the stock price movement by discovering the relationship and well diversified their investment portfolio in order to increase the return for them. Besides that, it is important for policy makers to discover the relationship between them as well because the policy makers need to year, this will influence the economic growth of our country because when our currency has depreciated, our purchasing power towards foreign goods will be reduced and this means that we need to use more money to purchase foreign goods compare to the past. In the world of gold price, it has reached to a new high price in this year, this might cause the people to invest in gold for hedging purposes instead of investing money in the stock market. Besides that, our stock market is being influenced by the US stock market and when the US market drop, usually by the next day Malaysia stock market will drop as well, this is mainly because previously our country currency is pegged to US dollar, so the US country economic can influence our country to a certain extent.

Methodology
Data are collected from January 2000 to December 2013 using monthly data; the period of this study is 14 years. There are total 1008 observations used in this study, out of total 1008 observations, there are 168 observations for each variable. Monthly data is adopted in study instead of quarterly data is to increase the accuracy of the finding as well as to avoid the problem of thin trading and price limits of a stock market. 3 tests are conducted such as unit root test, Johansen & Juselius co-integration test and granger causality test, each of them is testing in different perspective.

Unit Root Test
Unit root test is to tests whether a time series variable is non stationary using an autoregressive model and aimed at establishing the order of integration of each variable. If the variables are not stationary, then it can be proved that the standard assumptions for asymptotic analysis will not be valid. And it might lead to spurious regression. According to Seyed, Zamri and Lai (2011)

Johansen Co-integration Test
The Johansen & Juselius test is to determine whether the linear combination of the series have a long run equilibrium relationship or not, if there is co-integration relationship among these variables, a linear combination of the integrated series is stationary.

Granger Causality Tests
This study will investigate the causal relations between Malaysia stock market index and macroeconomic variables which mentioned early by using Granger Causality test which proposed by Granger (1969) and these variables might be used predict each other if the causal relationship existed. There are 2 steps in the Granger Causality testing. Firstly, stationary data is needed rather than non-stationary data, afterward, granger methodology which is sensitive to the lag length is used to test the stationary property of the data (Chong & Tai, 1999). If stock market can granger cause macroeconomic variable, so it will be very useful to policy maker in order to forecast the macroeconomic and investor in seek for higher return (Zukarnain & Sofian, 2012

Figure 1. Jarque-Bera Normality Test
The P-value for Jarque-Bera is 0.238855, it is > 0.01, so do not reject H 0 . Thus this shows that the error term is normally distributed and can continue to perform other tests on these data because it has been proven that the residual is in normality distributed.

Unit Root Test
This study has employed Augmented Dickey-Fuller and Phillips-Perron to test for stationary for each variable. The results are shows in Table 1.  The value in parenthesis refers to t-statistic. ***, ** and * indicate significant at 1%, 5% and 10% respectively.
Based on the above tables, all the variables are not significant, therefore we cannot reject H 0 and they are not stationary and contain unit root in both ADF test and PP test. However, when proceed to first difference, all the variables are significant at 1% and therefore reject H 0 and this can conclude that all the variables are stationary and do not contain root.

Johansen & Juselius Co-Integration Test
The samples are divided into 2 groups to test the co-integration relationship between local macroeconomic variables and Malaysia stock market as well as between foreign macroeconomic variables and Malaysia stock market. The results are show in Table 2. The results also show maximum eigen. value are significance at 5% in r=0, meaning that the co-integration relationship does exists among the Malaysia stock market and local macroeconomic variables, there is at least 1 macroeconomic that can influence Malaysia stock market in long run.

Granger Causality Test
It is used to check the causality relationship and determine the direction of this causality effect. Results

are show in below:
Causality test result among KLCI and foreign macroeconomic variables:

Conclusion
As a conclusion, the results from this study found that these variables are stationary at the first difference