Financial analytics for interlinking stock market and macroeconomic performance- post financial crisis 2008

Abstract

Financial analytics has been highly crucial in forecasting possible future economic scenarios. The relationship between a country’s macroeconomic indicators and its stock market has been extensively studied in the literature. Stock prices should be used as leading indications of future economic activity if they accurately reflect the underlying fundamentals. On the contrary, if economic activity follows stock price movement, the outcomes should be the opposite, i.e., economic activity should lead stock price movement. The paper attempts to make use of financial descriptive analytics to explore the interconnection between prominent macroeconomic indicators and stock market activity post ten years of financial crisis 2008. The study’s range is constrained to explore the aforementioned interconnection for the period from September’ 2008 to August’ 2018. The following factors have been found to be related over the long term: GDP, Production Index, Inflation, Exchange Rate, Money Supply, Imports, Exports, FDI, and Stock Market Returns. Shockingly FII has not shown any cointegrating equation. Also causality was observed between stock market and economic indicators. Impulse Response Function (IRF) and Variance Decomposition (VDC) techniques of VAR model are applied to decompose or fractionalize the variability caused by macroeconomic indicators on the BSE Sensex returns which has given some interesting results.

Keywords

Stock market financial analytics macroeconomic data

1. Introduction

The Indian stock market has witnessed turbulent changes in the last twenty years. With rising to the lifetime peak and then coming to an unexpected low, it has for sure given the testimony of its volatile nature. The domestic factors that include monetary policy changes, assembly elections and corporate scams play a significant role in stock market movements which is seemingly visible with markets touching new highs post elections and cabinet formation. We are no longer an isolated or insulated economy, with the global economic crisis hitting at its best to Indian stock market performance. The more recent factors have been the US trade policy, crude oil prices and fed rate that are being constantly linked to stock market performance. Advanced analytics techniques offer new insights into the problems of finance and economics.

The best ever indicator of stock market movement Sensex climbed to 39000 on 1 ${}^{\text{st}}$ April, 2019 exactly after 40 years of its formation. Since its formation it has always remained the gauge of economic health of financial markets. Given that it has achieved numerous milestones over the past 40 years, we can consider it a witness to the development of the Indian economy. Looking at this length of market journey, Sensex has crossed milestones, reached new lows but has never perfectly reacted to any market stimuli. It marked the level 21,000 in January’ 2008 and then nosedived to the level of 8,500 in December (down by 64 percent) making it the biggest fall ever in its history. This decline was attributed to the global sub-prime crisis, which began in 2008 and had an impact on Indian markets by October of that year. This volatile nature of the stock market which grew more prominent post 2008 created a lot of curiosity across academicians and researchers to investigate the nature and cause of this volatility and link it with other economic indicators. The paper attempts to discuss the interrelationship that existed between the stock returns and macroeconomic indicators for the 10 years period from 2008 to 2018.

2. Literature review

A lot of relevant research has been studied to explore the topic in depth. Few are discussed with respect to the variables studied and results obtained. Also the techniques which have been used play a significant role in deciding the genuineness of the relationships being explored.

A simple study was conducted by Ilahi et al. (2015) of Karachi stock market to explore the interdependency of stock market returns and other macroeconomic indicators notably, the rates of inflation, interest, and currency exchange. Using a multiple regression analysis it was concluded that all the three variables had insignificant relationship with stock returns. A study was performed by Fang and Miller (2002) during the late 1990s Korean financial crisis to look into the impact of daily currency depreciation on returns on the Korean stock market. It was found that the Korean stock market and the foreign currency market are causally related in both directions.

Purna Chandra (2006) performed a similar task in analysing the dynamic relationship between the stock price and exchange rate in India. The long run, short run, and causative aspects of the relationship are all considered. The empirical results suggested that there exists a long run relationship between exchange rates and stock prices but no such signs exist in the short run. The study found that unidirectional link flow from exchange rate to stock prices in the long run.

Ahmed and Imam (2007) investigated using a cointegration test, if present economic activity in Bangladesh can account for stock market returns over the long term. Additionally, it was discovered that there was no cointegration between fundamental macroeconomic parameters and the increase of stock market returns, which was assumed to be the result of Bangladesh’s small and shallow nascent stock market. It was anticipated that the market return might be affected by changes in interest rates or the growth rate of T-bills. It was discovered that a change in interest rates under Granger leads to unidirectional stock market returns, suggesting that the stock market index is not a leading indicator for the change in interest rates as an economic variable. This resulted in evidence of an informationally inefficient market.

Deb and Mukherjee (2008) studied the link between the rise of the Indian stock market and economic expansion over the past ten years. The study period chosen was just over a decade of 1996–2007. The presence of a causal relationship between development in the real sector and the stock market in India was explored and the exact direction of the relationship was understood. It was observed that there is bidirectional causality between real GDP growth rate and real market capitalization ratio.

The causal relationship between stock indices and the key economic variables was investigated by Shahid Ahmed (2008). For the years 1995 to 2007, the analysis included the following variables: the index of industrial production, exports, foreign direct investment, money supply, exchange rate, interest rate, NSE Nifty, and BSE Sensex. The study examined the long- and short-term causal relationships and found that while changes in the BSE Sensex appear to affect these variables, changes in the NSE had no effect on exchange rates and IIP. Additionally, NSE Nifty influences exchange rates, IIP, and money supply in the near term, whereas interest rates and FDI influence NSE Nifty, which is followed by the similar outcome for Sensex.

Rakesh Kumar (2013) examined the different macroeconomic to establish the variables that affect the performance of the Indian stock market. The different economic factors selected for the study purpose were money supply, consumer price index, Gold, crude oil, foreign exchange reserves, FIIs, FDI, call money rate, balance of trade, repo rate, foreign exchange rate, and industrial growth rate. It was also observed that the industrial performance is extremely passively correlated with the stock market performance as measured by growth patterns and performance of the stock market is actively related to the macroeconomic environment.

3. Data and methodology

To inter relate the stock indices to the different macroeconomic variables selected, the monthly data of the variables for the period September’ 2008 to August 2018 have been collected. So the total number of observations for each variable for the study period is 120. The data has been collected for the variables Gross Domestic Product (GDP), Production Index (IIP), Inflation (WPI), Exchange Rate (EXCRATE), Money Supply (MS), Imports (IMP), Exports (EXP), FDI, FII, Gold Prices (GP) and BSE Sensex.

The descriptive statistics are presented in Table 1. The percentage change in the monthly closing prices is used to calculate all returns. The mean value is highest in case of FDI and lowest in case of EXCRATE and WPI. FII has got a negative mean. The standard deviation of all the variables shows that FII, FDI and IIP have got higher volatility whereas WPI, EXCRATE and MS are less volatile. The frequency distribution places a larger chance on returns near zero as well as extremely high positive and negative returns because the kurtosis for all the variables is greater than 3. The statistic of Jarque-Bera along with its probability shows that maximum variables are not normally distributed.

Table 1
Descriptive statistics of macroeconomic variables and BSE Sensex for the period September 2008 to August 2018

Statistics	Variables
	GDP	IIP	WPI	EXCRATE	MS	IMP	EXP	FDI	FII	GP	Sensex
Mean	1.5130	7.3691	0.5161	0.1358	1.3170	2.2238	2.4589	32.462	$-$ 68.580	1.5473	1.7689
Median	0.9702	0.9425	0.4766	$-$ 0.1133	1.0742	2.2169	1.6579	4.8813	$-$ 37.380	1.4258	2.6100
Maximum	14.940	848.29	2.5785	6.7861	5.9009	27.551	29.590	876.37	602.94	14.367	21.296
Minimum	$-$ 15.674	$-$ 89.785	$-$ 1.8912	$-$ 4.2652	$-$ 0.5056	$-$ 17.786	$-$ 29.971	$-$ 76.770	$-$ 1827.1	$-$ 10.201	$-$ 24.336
Std. Dev.	6.4296	78.411	0.6619	1.9940	1.0907	9.7066	11.734	124.09	290.64	3.9589	6.4595
Skewness	$-$ 0.1392	10.479	$-$ 0.0998	0.7438	1.2658	0.2771	0.0404	4.0814	$-$ 2.4163	0.1814	$-$ 0.4882
Kurtosis	3.2136	113.07	4.6735	4.3448	5.7105	2.4530	3.2261	24.383	14.966	3.8782	5.1832
Jarque-Bera	0.6111	62260.54	14.084	19.941	68.211	3.0066	0.2861	2597.5	825.87	4.4774	28.361
Probability	0.7367	0.0000	0.0008	0.0000	0.0000	0.2223	0.8666	0.0000	0.0000	0.1065	0.0000
Observations	119	119	119	119	119	119	119	119	119	119	119

The Augmented Dickey-Fuller test and Philips-Perron test is applied to test whether the series have unit root or not i.e. to test the stationarity of the series which is a prime condition to apply any econometric model.

Table 2

Results of augmented Dickey-Fuller test and Phillips-Perron test

Variables	Augmented Dickey-Fuller test				Phillips-Perron test
	With intercept		With trend and intercept		With intercept		With trend and intercept
GDP	$-$ 2.	5733	$-$ 2.	5896	$-$ 25.	3335 ${}^{*}$	$-$ 25.	1851 ${}^{*}$
D(GDP)	$-$ 11.	7836 ${}^{*}$	$-$ 11.	7234 ${}^{*}$	–		–
IIP	$-$ 11.	9648 ${}^{*}$	$-$ 11.	9469 ${}^{*}$	$-$ 11.	9759 ${}^{*}$	$-$ 11.	9598 ${}^{*}$
WPI	$-$ 5.	7558 ${}^{*}$	$-$ 5.	8417 ${}^{*}$	$-$ 8.	5478 ${}^{*}$	$-$ 8.	5467 ${}^{*}$
EXCRATE	$-$ 7.	5825 ${}^{*}$	$-$ 7.	6734 ${}^{*}$	$-$ 7.	5133 ${}^{*}$	$-$ 7.	5788 ${}^{*}$
MS	$-$ 11.	2639 ${}^{*}$	$-$ 11.	2256 ${}^{*}$	$-$ 11.	5016 ${}^{*}$	$-$ 11.	4797 ${}^{*}$
IMP	$-$ 15.	0105 ${}^{*}$	$-$ 14.	9813 ${}^{*}$	$-$ 14.	9510 ${}^{*}$	$-$ 14.	9446 ${}^{*}$
EXP	$-$ 6.	6579 ${}^{*}$	$-$ 6.	6702 ${}^{*}$	$-$ 22.	8995 ${}^{*}$	$-$ 23.	1056 ${}^{*}$
FDI	$-$ 12.	9153 ${}^{*}$	$-$ 12.	8660 ${}^{*}$	$-$ 12.	9282 ${}^{*}$	$-$ 12.	8785 ${}^{*}$
FII	$-$ 10.	2883 ${}^{*}$	$-$ 10.	3260 ${}^{*}$	$-$ 10.	3042 ${}^{*}$	$-$ 10.	3359 ${}^{*}$
GP	$-$ 10.	7126 ${}^{*}$	$-$ 10.	6823 ${}^{*}$	$-$ 10.	7133 ${}^{*}$	$-$ 10.	6839 ${}^{*}$
SENSEX	$-$ 4.	4274 ${}^{*}$	$-$ 4.	5677 ${}^{*}$	$-$ 7.	3584 ${}^{*}$	$-$ 7.	5487 ${}^{*}$
Critical values
Level of significance (1%)	$-$ 3.	4865	$-$ 4.	0376	$-$ 3.	4865	$-$ 4.	0376
Level of significance (5%)	$-$ 2.	8860	$-$ 3.	4483	$-$ 2.	8860	$-$ 3.	4483
Level of significance (10%)	$-$ 2.	5799	$-$ 3.	1493	$-$ 2.	5799	$-$ 3.	1493

${}^{*}$ Null hypothesis is rejected at 1% level of significance.

The results of the ADF and PP stationarity test of the concerned time series are shown in Table 2. It can be seen from the table that the absolute value of the test statistics for both the unit root test for the total number of variables are higher than the threshold. So the null hypothesis is rejected and the data is found to be stationary. We accept the alternative hypothesis that the series of all the selected variables are stationary and have no unit root. The value of the test statistic for GDP in Augmented Dickey-Fuller test is $-$ 2.5733 and $-$ 2.5896 which is less than the critical value even at 10 percent level of significance. At the first difference, it is discovered that the GDP series is stationary. As a result, it is now possible to use the VAR Granger causality test and multiple regression model, which both need stationary data to avoid producing false results.

To analyse the relationship, cointegration test is used between different macroeconomic variables and stock price indexes. Natural logs are created from the variables which are found stationary at first difference. Cointegration analysis has to be used to determine if there is any long-term equilibrium relationship for non-stationary series. Regression analysis using non-stationary series introduces difficulties for statistical inference. This notion gives an explanation for the long-term link between the two variables and how to test for cointegration between two variables because two variables that are cointegrated, on average, would not drift apart over time.

A Vector AutoRegression (VAR) methodology has been adopted where several endogenous variables are taken into account, the lagged values of each variable and all other endogenous variables in the model are used to explain each variable. The VAR model takes into account several variables at once, but the pairwise Granger’s causality test only evaluates the cause and effect relationship between the two variables at a time. The lagged value coefficients of the first variable must be significant in the VAR if one variable “Granger causes” another. It is challenging to determine which set of factors have a meaningful impact on each of the endogenous variable and which do not because VAR incorporates multiple lags of variables. This problem is addressed with the VAR Granger Causality or Block Exogeneity Wald test, which limits all lags of a certain variable to zero.

To decompose or fractionalize the variability caused by macroeconomic indicators on the BSE Sensex returns, the methods of Variance Decomposition (VDC) and Impulse Response Function (IRF) of the VAR model have been used. While the VDC breaks down the variance in an endogenous variable into shocks to the VAR, the IRF tracks the result of a change of one endogenous variable on the other variables in a VAR.

IRF is a technique that has traced out the responsiveness of the BSE Sensex return to the shocks or innovations or impulses given to the key macroeconomic indicators. The responsiveness of all the variables is presented irrespective of their causality. Variance decomposition will reveal the quantum of changes in BSE Sensex explained by it and the portion explained by other key macroeconomic variables viz. GDP, IIP, EXP, FDI and GP. As a result, the changes in the BSE Sensex return have been broken down into the individual shocks that make up the VAR model, and information about the relative significance of each random innovation in changing the macroeconomic variables in the VAR model has been provided.

Table 3

Result of cointegration test between BSE Sensex and macroeconomic variables

Variables	Hypothesized no. of CE’s	Trace statistics	Prob.	Max-Eigen statistics	Prob.
GDP	None ${}^{*}$	52.3459	0.0000	33.6564	0.0000
	Atmost 1 ${}^{*}$	18.6894	0.0000	18.6894	0.0000
IIP	None ${}^{*}$	83.9856	0.0000	65.8361	0.0000
	Atmost 1 ${}^{*}$	18.1474	0.0000	18.1474	0.0000
WPI	None ${}^{*}$	48.3630	0.0000	28.1441	0.0000
	Atmost 1 ${}^{*}$	20.2188	0.0000	20.2188	0.0000
EXCRATE	None ${}^{*}$	50.8960	0.0000	31.4838	0.0000
	Atmost 1 ${}^{*}$	19.4122	0.0000	19.4122	0.0000
MS	None ${}^{*}$	53.9007	0.0000	35.9273	0.0000
	Atmost 1 ${}^{*}$	17.9734	0.0000	17.9734	0.0000
IMP	None ${}^{*}$	48.8022	0.0000	29.7944	0.0001
	Atmost 1 ${}^{*}$	19.0077	0.0000	19.0077	0.0000
EXP	None ${}^{*}$	48.6940	0.0000	31.0067	0.0001
	Atmost 1 ${}^{*}$	17.6872	0.0000	17.6872	0.0000
FDI	None ${}^{*}$	61.4964	0.0000	43.2884	0.0000
	Atmost 1 ${}^{*}$	18.2080	0.0000	18.2080	0.0000
FII	None	13.5897	0.0949	13.3902	0.0684
	Atmost 1	0.1995	0.6551	0.1995	0.6551
GP	None ${}^{*}$	39.6992	0.0000	22.5321	0.0020
	Atmost 1 ${}^{*}$	17.1671	0.0000	17.1671	0.0000
Critical values at 5 per cent
None ${}^{*}$		15.4947		14.2646
Atmost 1 ${}^{*}$		3.8414		3.8414

${}^{*}$ Null Hypothesis is rejected at 5% level of significance.

Table 4

VAR granger causality test results

Null hypothesis	Wald-statistics		$P$ -value
GDP does not lead to BSE Sensex	21.	1970 ${}^{*}$	0.0066
BSE Sensex does not lead to GDP	22.	5899 ${}^{*}$	0.0039
IIP does not lead to BSE Sensex	16.	3630 ${}^{*}$	0.0375
BSE Sensex does not lead to IIP	4.	6010	0.7992
WPI does not lead to BSE Sensex	5.	0779	0.7492
BSE Sensex does not lead to WPI	6.	8987	0.5476
EXCRATE does not lead to BSE Sensex	10.	1077	0.2575
BSE Sensex does not lead to EXCRATE	9.	0911	0.3347
MS does not lead to BSE Sensex	13.	3349	0.1008
BSE Sensex does not lead to MS	3.	0801	0.9292
IMP does not lead to BSE Sensex	10.	8834	0.2084
BSE Sensex does not lead to IMP	6.	6498	0.5748
EXP does not lead to BSE Sensex	22.	2289 ${}^{*}$	0.0045
BSE Sensex does not lead to EXP	6.	4589	0.5960
FDI does not lead to BSE Sensex	26.	7198 ${}^{*}$	0.0008
BSE Sensex does not lead to FDI	18.	4008 ${}^{*}$	0.0184
GP does not lead to BSE Sensex	14.	8512 ${}^{*}$	0.0064
BSE Sensex does not lead to GP	13.	3177	0.1014

${}^{*}$ Null Hypothesis is rejected at 5% level of significance.

Table 5

Results of variance decomposition of BSE Sensex, GDP, IIP, EXP, FDI and GP

Variance decomposition	Lags	Sensex		GDP
Sensex	2	99.	929	0.	071
	5	99.	922	0.	078
	10	99.	581	0.	419
GDP	2	0.	444	99.	556
	5	4.	000	96.	000
	10	18.	228	81.	772
Variance decomposition	Lags	Sensex		IIP
Sensex	2	99.	547	0.	453
	5	99.	735	0.	265
	10	99.	757	0.	243
IIP	2	0.	187	99.	813
	5	2.	058	97.	942
	10	7.	148	92.	852
Variance decomposition	Lags	Sensex		EXP
Sensex	2	99.	206	0.	794
	5	99.	435	0.	565
	10	89.	843	10.	157
EXP	2	0.	761	99.	239
	5	4.	140	95.	860
	10	18.	386	81.	614
Variance decomposition	Lags	Sensex		FDI
Sensex	2	99.	768	0.	232
	5	99.	828	0.	172
	10	99.	686	0.	314
FDI	2	1.	019	98.	981
	5	3.	004	96.	996
	10	11.	002	88.	998
Variance decomposition	Lags	Sensex		GP
Sensex	2	99.	396	0.	604
	5	99.	532	0.	468
	10	99.	206	0.	794
GP	2	0.	209	99.	791
	5	0.	933	99.	067
	10	3.	399	96.	601

4. Results and findings

H ${}_{1}$ : There are cointegrating equations between macroeconomic factors and stock indexes.

Table 3 presents the result of the Johansen Cointegration test used to examine the long run relationship between Sensex and the macroeconomic variables selected viz. GDP, IIP, WPI, EXCRATE, MS IMP, EXP, FDI, FII and GP. The series is stationary at level but becomes stationary at first difference. The number of lags applied in the cointegration analysis is 8 as per Akaike Information Criteria. The two statistics examined were Trace statistics and Max-Eigen statistics. Null hypothesis is rejected at 5 percent level of significance for all the variables except FII which means that for all the macroeconomic factors excluding the Foreign Investments (FII), there existed two cointegrating equations with respect to BSE Sensex at 5% level. For FII the value of the trace statistics and max-eigen statistics is 13.5897 and 13.3092 which is less than the critical values of 15.4947 and 14.2646 at 5% level. The test statistics value of the rest of the variables is greater than the critical value indicating the presence of cointegrating equations.

Table 4 shows the result of the Granger Causality test used to determine the flow of any causality running from the macroeconomic variables to stock indices and vice versa.

H ${}_{2}$ : Macroeconomic variables cause BSE Sensex.

H ${}_{3}$ : BSE Sensex causes macroeconomic variables.

Both the hypotheses are tested for the entire selected macroeconomic variables and the impact of the variables on the stock performance is analyzed through unidirectional and bidirectional causality.

The impulse responses of BSE Sensex return to all the macroeconomic variables are also included in the VAR model. It has been noted that the BSE Sensex reacts favourably to any impulses introduced from IIP, IMP and FDI and has responded in an opposite direction to EXCRATE and EXP. It has given a mix of positive and negative responses to GDP, WPI, MS and GP.

The variance decomposition for BSE Sensex and the other macroeconomic variables having one way or two way causality is being presented in Table 5. It is seen that while GDP explains 0.40 percent of the variation in BSE Sensex, BSE Sensex demonstrate around 18 percent of the forecast variance in GDP at 10 lags. Similar shocking results can also be seen in the case of IIP where BSE Sensex clarify 7 percent of the variance in IIP and IIP clarify only 0.24 percent of the variation in BSE Sensex. In the case of EXP and BSE Sensex, they explain 10 and 18 percent of variance of each other at 10 periods. BSE Sensex explains around 11 percent of the variations in FDI while FDI explains about 0.3 percent of BSE Sensex. Lastly at 10 periods lag BSE Sensex explains about 3 percent of GP while GP explains only 0.8 percent of variance in BSE Sensex.

5. Conclusion

There is a long run relationship between GDP, IIP, WPI, EXCRATE, MS, IMP, EXP, FDI and BSE Sensex returns. It was shocking to find out that FII whose major inflows and outflows has an immediate impact on stock market movements has not shown any cointegrating equation with BSE Sensex.

There is a short run bidirectional causality running between GDP and BSE Sensex, and FDI and BSE Sensex while one-way causality was found between IIP, EXP, GP and BSE Sensex. In the short run GDP and BSE Sensex, and FDI and BSE Sensex cause each other while IIP, EXP and GP cause BSE Sensex in the short run.

BSE Sensex shows a positive change or responsiveness towards any shock brought to the IIP, IMP and FDI while it gives a negative response with respect to variables EXCRATE and EXP. A kind of mixed response was observed on the BSE Sensex with respect to shocks introduced in GDP, WPI, MS and GP.

BSE Sensex explains a greater portion of the variability of the variables GDP, IIP, EXP, FDI and GP than the variables explaining variance of BSE Sensex. Moreover, BSE Sensex and other variables like IIP, EXP and GP explain a bigger component of its own forecast variance than forecast given by other variables.

The significant long-term correlation between the stock market and macroeconomic indicators performance calls for a detailed analysis of the macroeconomic factors of a country before making any stock market investment for a long term period. Any long term investment decision should be based on the study of the macro fundamentals of the country. The possible causes of FII not being into any long run association with stock market returns can be explored further as FII’s activities in the short run are being reflected immediately by the stock market.

References

Ahmed and Imam (2007). Macroeconomic factors and bangladesh stock market: Impact analysis through co integration approach. International Review of Business Research Papers, 3(5), 21-35.

Ahuja

H.L.

(2008). Macroeconomic Theory and Policy, S. Chand & Company Pvt, Ltd., New Delhi.

Datt

, & Mahajan

(2013). Indian Economy, S. Chand & Company Pvt, Ltd., New Delhi.

Fang

, & Miler

S.M.

(2002). Currency Depreciation and Korean Stock Market Performance during the Asian Financial Crisis. Retrieved from http//www.Econ.uconn,edu/working/2002.

Gujarati

D.N.

(2004). Basic Econometrics, Tata McGraw Hill, New Delhi.

Gupta

G.S.

(2006). Macroeconomics Theory and Applications, Tata McGraw Hill Publishing Co. Ltd., New Delhi.

Ilahi

Ali

, & Jamil

(2015). Impact of Macroeconomic Variables on Stock Market Returns: A Case of Karachi Stock Exchange. Retrieved from SSRN Electronic Journal. 10.2139/ssrn.2583401.

Maccarrone, Morelli, Spadaccini (2021). GDP forecasting: Machine learning, linear or autoregression? Frontiers in Artificial Intelligence, 4. doi: 10.3389/frai.2021.757864.

Purna

C.P.

(2006). The dynamic relationship between the stock price and exchange rate in India. The ICFAI Journal of Monetary Economics, IV(3), 26-36.

10.

Shahid

(2008). Aggregate economic variables and stock markets in India. International Research Journal of Finance and Economics, Issue 14(2008), 143-163.

11.

Soumya

G.D.

, & Jaydeep

(2008). Does Stock Market Development Cause Economic Growth? A Time Series Analysis for Indian Economy. International Research Journal of Finance and Economics, Issue 21, 142-149.

12.

Wan Mansor Wan Mahmood and Nazihah Mohd Dinniah (2009). Stock Returns and Macroeconomic Variables-Evidence from Six Asian-Pacific Countries. International Research Journal of Finance and Economics, Issue 30(2009), 154-164.