INTRODUCTION:

Indian stock market is working from several years ago and now developed as crucial market for Indian economy but it has opened to foreign institutional investors in September 1992 since then significant amount of investment from foreigners in the form of foreign institutional investor (FII) investment in equities is coming in India. During 1992-93, a net FII flow in the Indian equity market participation was only about Rs.13.40 crore and afterwards it was amplified and reached a level of Rs. 45,881 Crore in 2004-05. FII flows in India drastically declined to Rs. 43,737.60 Cr in 2011 – 12 from Rs.1, 10, 120.80 Cr in 2010-11. FIIs investment is volatile by nature and depend on market change so FIIs flows might be positive of negative.

It is an assumption that asset prices in efficient market reflect the information possessed by informed investors and noise traders (Black, 1986). The noise traders provide liquidity and volatility to the market, whereas the informed investors search for mispriced stocks and invest in them. Informed investors reduce volatility and provide stability of stock prices but are unable to eliminate the effects of noise traders (Thaler, 1999). Noise traders cause volatility in the asset prices as they move the asset price away from their fundamentals value (De Long et al., 1990). Positive feedback trading is one of the forms of noise trading where investors believe that the current sentiments in the market would continue.

As we have studied several literatures on foreign institutional investment and nifty index and most of literatures used traditional Johansen co-integration test for examine the co-integration between the variables. But in our study we have applied ARDL model which gives more appropriate result than traditional approach. Under this model optimum lag selection and best fit model is possible which was not in traditional approach.

REVIEW OF LITERATURE:

Kumari, J and Mahakud, J (2015), studied the theoretical linkage between stock market volatility and macroeconomic volatility in emerging Indian stock market. The study covers the period on and from July 1996 to March 2013. Unlike the previous studies, the study investigated the issue with two stage estimation techniques. Conditional volatility is extracted by employing univariate autoregressive conditional heteroskedasticity models. Further, multivariate VAR model along with impulse response function, block exogeneity and variance decomposition are carried out to analyze the relationship between stock market volatility and macroeconomic volatility. Data on macroeconomic variables namely output, foreign institutional investments, exchange rate, short term and long-term interest rates, broad money supply, inflation and stock market indices BSE Sensex and NSE Nifty are used for analysis. The findings suggest a linkage between macroeconomic volatility and equity market volatility (Kumari and Mahakud). Dr. Ranjan Dasgupta conducted a study on stock market return and macroeconomic variables and found that return on stocks, domestic macroeconomic variables, and India-specific stock market-driven factors have been the most dominant determinants of FIIs (Trivedi P). Another study exposed that FII inflows to India are fundamentally determined by exchange rate, domestic inflation, domestic equity market returns, returns and risk allied with US equity market (Srinivasan P). A study was conducted by Kaur and sharanjit in 2010 on the determinants of Foreign Institutional Investors’ (FIIs) investment in India. The study depict that Returns on Indian stock market have positive significant impact while return of US stock market have no significant influence on FIIs investment to India. Bodla B S and Kumar A (2009) examine the FIIs correlation with economic variables for a period of 15 years from 1993 to 2007and the correlation was examined by using Engle Granger Causality test. The study represent that the net investment made in Indian stock market by the foreign institutional investors is important factor of market capitalization but in case of trading volume, FIIs investment turned as a result of trading volume.

Research Gap:

Previous study confirmed that most of the study used Johansen Co-integration for examine the long run relationship between the variables on short period data but we have applied ARDL model on long period monthly data for analyse the short run and long run relationship which is more appropriate than traditional test. Because the ARDL test check the stability and best fit model.

OBJECTIVE OF THE STUDY:

1. To examine the Short - run and long - run Co-integration between FIIs and Nifty Indices.

2. To generate the best fitted model for the series.

3. To check the stability of the best fitted model.

RESEARCH METHODOLOGY:

Data and Period of study:

Econometric Tools- The methodology employed for examine the inter-linkage among exogenous and endogenous variables using co-integration test for long run relationship and causality test for short run relationship. Here we have applied following process and techniques for attaining above objectives-

Lomnicki–Jarque–Bera Test for Non-normality

Unit Root and Stationarity Tests -

Augmented Dickey–Fuller Tests (ADF)

Kwiatkowski, Phillips, Schmidt, and Shin (KPSS) Test

Lag Length Selection

ARDL Co-integration (Short and Long-run relationship)

Residual Diagnostic

Jarqua- Bera Normality Test

Heteroscedasticity Test

Serial Correlation (LM test)

Stability Test

CUSUM Test

Ramsay RESET Test

Table 2 shows result of Stationarity test using ADF test and KPSS test and it confirmed that FIIs is coming stationary at level and nifty price at first difference which affirm degree of integration of the variable. So we can say one variable is I (0) and another I (1) series and we can apply ARDL model for investigating co-integration between the variables.

Auto Regressive Distribution Lag (ARDL) Model:

(Short run and long run co-integration)

Sample (adjusted): 2000M03 2022M12

Model selection method: Schwarz criterion (SIC)

Selected Model: ARDL (2, 1)

Table-3

Variable	Coefficient	Std. Error	t-Statistic	Prob.*
FIIS__CR__(-1)	0.150932	0.049496	3.049394	0.0025
FIIS__CR__(-2)	0.145501	0.048339	3.010026	0.0029
NIFTY	19.08728	1.471596	12.97046	0.0000
NIFTY(-1)	-19.39810	1.487819	-13.03794	0.0000
C	3451.517	867.7995	3.977320	0.0001

Source: Own Calculation

Table 3 shows the description of auto regressive distributive lag model which has been applied on two variable i.e. Foreign Institutional Investment and Nifty indices. Here Foreign Institutional investment is taken as endogenous variable and nifty indices is exogenous for applying ARDL model.

The basic properties of the model is that the data should be I (1) and I (0) series so that we can make further processing for analysis. Table –02 depicts that there is significant positive relationship between FIIs and Nifty indices in short run and significant negative relationship in long run. As a change in nifty by 1 percent accelerates positive change in FII investment by 19.0872 Cr in short run and declining the nifty index affect FIIs investment negatively in long run.

Selection of Best Fit Model Using Schwarz Criteria:

For selection of appropriate model for applying ARDL we have used Schwarz criteria, as per the properties of selection of model the lowest spike in the diagram will be best fitted model so figure 2 shows best model for applying ARDL that is ARDL (2, 1).

Long Run Bound Test:

Null Hypothesis: No levels relationship

Table -4

F-Bound Test
Test Statistic	Value	Sign	I(0)	I(1)
F-statistic	57.58212	10%	3.02	3.51
k	1	5%	3.62	4.16
		2.5%	4.18	4.79
		1%	4.94	5.58

Source: Own Calculation

In the next step, we performed ARDL bounds test to examine the existence of cointegration. The bounds test approach on all six alternative versions of Wagner’s hypothesis has been used to examine long-run relationship between the variables. To know the appropriate lag length of the variables in ARDL model, we used the AIC and SBIC criteria. Order of the variables in different versions is presented in Table 4. As per property of ARDL F - statistics should come more than table value, here F-statistics (57.58212) is coming more that lower bound (3.62) and upper bound (4.16) which show significant short and long run co-integration between the variables. So, as nifty today’s and lag price change it affects the investment decision of investors.

Error Correction Model:

Table 5

ECM Regression
Case 2: Restricted Constant and No Trend
Variable	Coefficient	Std. Error	t-Statistic	Prob.
D(FIIS__CR__(-1))	-0.145501	0.047536	-3.060827	0.0024
D(NIFTY)	19.08728	1.443977	13.21855	0.0000
CointEq(-1)*	-0.703567	0.053333	-13.19207	0.0000

Source: Own Calculation

This model show combined effect of short and long run co-integration. Table 5 depicts short run effect is positive and long run effect is negative. In short as change in nifty by one it will make change by 19.08728 Cr in FIIs and in long run, any change in nifty would affect FIIs negatively by -0.703567. In short run investors go for momentum trading and in long run they go for value trading.

Wald Test- Long run relationship:

Null Hypothesis: C(3)=C(4)=0

Table 6

Wald Test:
Test Statistic	Value	df	Probability
F-statistic	85.09977	(2, 269)	0.0000
Chi-square	170.1995	2	0.0000

Source: Own Calculation

We can also check long run relation by this test and this is showing that there is significant long run relationship between variables because F-statistic and Chi-square value is coming so high and significant.

· Residual Diagnostic Test:

In residual diagnostic test there are four type of test we have used. Normality test used for checking the distribution of data in a series, ARCH LM test used for examine the serial correlation problem and for checking the stability of model Ramsey RESET test has been used. Following table depicts the results of these entire tests.

Table 7

Diagnostic Test
Test	FIIs & Nifty
Normality Test	0.0000
LM Test	1.154703(0.3167)
RESET Test	0.3885(0.7460)

Source: Own Calculation

Table 7 represents normality, ARCH LM and Ramsey RESET Test that analyse stability of model. Here we can understand that the model is statistically significant and stable because as per the property of LM and RESET test the p-value should be greater than 0.05 and value of both ARCH LM and Ramsey RESET test are coming greater than 0.05. So we can say model is stable and appropriate future investment decision.

RECURRSIVE DIAGNOSTIC:

This test analyse the stability of model whether the model is statistically fit or not. For this, following test have been applied which are as follow-

CUSUM test- Figure 3

FINDING AND CONCLUSION:

As we know India is a growing economy in the world and there is high opportunity to invest in Indian market. In the last few decades Indian stock market has given high return to the investors. So, in this decade foreign institutional investors have invested huge amount of money in Indian stock market. The study confirmed that FIIs and nifty price significantly co-integrated to each other in short run as well as long run. A change in nifty indices by 1 percent accelerates positive change in FII investment by 19.0872 Cr in short run and declining the nifty index affect FIIs investment negatively in long run which means investors go for momentum trading in short and value trading in long run.

In short run, market is so much volatile that why it gives high level of return to the investors so risk takers should go for momentum trading and the investors who wants to enhance their capital should go value trading or capital appreciation in long run.

Declaration of Conflicting Interests:

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

REFERENCE:

1. Ibrahim, M.H. and Aziz, H., “Macroeconomic variables and the Malaysian equity market: a view through rolling subsamples”, Journal of Economic Studies, (2003) , Vol. 30 No. 1, pp. 6-27,

2. Jalil, A., Mahmood, T. and Idrees, M., “Tourism–growth nexus in Pakistan: evidence from ARDL bounds tests”, Economic Modelling, (2013) Vol. 35, pp. 185-191.

3. Kuosmanen, P., Nabulsi, N. and Vataja, J, “Financial variables and economic activity in the Nordic countries”, International Review of Economics and Finance, (2015) Vol. 37, pp. 368-379.

4. Maio, P. and Philip, D. “Macro variables and the components of stock returns”, Journal of Empirical Finance, (2015) Vol. 33, pp. 287-308.

5. Malik, I.A. and Amjad, S. “Foreign direct investment and stock market development in Pakistan”, Journal of International Trade Law and Policy, (2013), Vol. 12 No. 3, pp. 226-242.

6. Malkiel, B.G. “The capital formation problem in the United States”, The Journal of Finance, (1979) Vol. 34 No. 2, pp. 291-306.

7. Marshall, D.A. “Inflation and asset returns in a monetary economy”, The Journal of Finance, ,(1992) Vol. 47 No. 4, pp. 1315-1342

8. Singh, P. “An empirical relationship between selected Indian stock market indices and macroeconomic indicators”, International Journal of Research in Business Management, (2014), Vol. 2 No. 9, pp. 81-92

9. Pesaran, M.H., Shin, Y. and Smith, R.J. “Bounds testing approaches to the analysis of level relationships”, Journal of Applied Econometrics, (2001) Vol. 16 No. 3, pp. 289-326,

10. Gay, R.D. “Effect of macroeconomic variables on stock market returns for four emerging economies: Brazil, Russia, India, and China”, International Business and Economics Research Journal, (2016) Vol. 15 No. 3, p. 119.

Received on 16.03.2023 Modified on 12.04.2023

Asian Journal of Management. 2023;14(2):141-145.

DOI: 10.52711/2321-5763.2023.00023

Variables	Mean	Std. dev.	Skewness	Kurtosis	Jarque–Bera	Prob.
FIIs	3708.805	10818.78	-0.091242	6.545838	144.9721*	0.0000
Nifty	6397.115	4551.842	0.845400	3.021298	32.88145*	0.0000

Unit Root Test for the Log Value of both Variables (with trend and Intercept)
	Augmented Dickey–Fuller Tests		Kwiatkowski, Phillips, Schmidt, and Shin (KPSS) Test
Variables	Stationarity at level	Stationarity at 1^st Difference	Stationarity at level	Stationarity at 1^st Difference
FIIs	0.0000*	0.0000**	0.137395*	0.104623**
Nifty	0.6863	0.0000**	0.324192	0.042051**