Date of submission: January 9, 2021; date of acceptance: March 4, 2021. * Contact information: v.kumariipmb@gmail.com, Indian Institute of Plantation Management Bengaluru, Jnanabharathi campus, Mallathalli – Post, Bangalore 560056, India, phone: +91-9844361528; ORCID ID: https://orcid.org/0000-0003-2319-2132. Copernican Journal of Finance & Accounting e-ISSN 2300-3065 p-ISSN 2300-12402020, volume 9, issue 4 Vijayakumar, A.N. (2021). Relativity of Indian Stock Market with Exchange Rate, Gold and Crude Oil. Copernican Journal of Finance & Accounting, 9(4), 101–118. http://dx.doi.org/10.12775/ CJFA.2020.024 a.n. vIjayakumar* Indian Institute of Plantation Management Bengaluru relatIvIty of IndIan stock market wIth exchange rate, gold and crude oIl Keywords: co-integration, crude oil, exchange rate, gold, stock market. J E L Classification: G15. Abstract: Stock market return is a motivating factor to investor in investment and portfolio decisions. Markets attracts domestic and foreign investments in anticipation of higher returns considering several parameters. These returns are inf luenced with economic, taxation and geo-political factors. Investment decision at market discounts with f luctuations of oil, exchange rate and gold. India being the largest consumer, the demand for crude oil and gold has been increasing and lead to higher import bill im- pacting f luctuations on exchange rate (USD-INR). Investor’s investment decisions at market discounts with volatility of oil, exchange rate and gold. This study with cau- sal research method using 25 years of data administered Johansen co-integration and Vector Error Correction Model to explore the relative impact of exchange rate, crude oil and gold on Indian stock market. The study finds presence of long run relationship of exchange rate, gold and crude oil with market returns and absence of short run rela- tionship. The findings shall facilitate in understanding the impact of f luctuations and investment decisions to benefit from Indian stock market. A.N. Vijayakumar102  Introduction Stock market of India is one of the leading Asian markets considering the rep- utation of market regulations and returns. It has been inf luenced by social, political and economic factors such as inf lation rate, exchange rates, f low of foreign investments, political stability, growth of gross domestic product, li- quidity and government policies of India and other inf luencing countries. Gold and Crude oil are the largest traded commodities and used as an important fi- nancial asset in portfolio management by investors and fund managers. Crude oil, apart from using as a fuel, is also an input for wide gamut of industries and subsequent price movement’s inf luences related prices of lubricants, fertiliz- ers, transportation and petrochemicals. Hence, prices of oil are a major concern to all stakeholders and economies of the world. It is also one of the inf luencing factors causing instability on stock market and its returns. India is one among the largest importing country of crude oil imported 82.1% of total consumption during the year 2016-17 (Dalei, Roy & Gupta, 2017). International transactions of settlement are largely through US dollars, hence, the greater demand for oil leads to def lation of Indian currency (Jain & Biswal, 2016). India is one of the largest jewellery market in the world consumes gold as an input to meet domestic and international processing requirements and is expected to increase around 33% by 2021. Gold in India is largely imported hence; valuation of Indian currency shall weaken against US dollar resulting appreciation of its value in terms of Indian rupee. The impact of rise in oil pric- es, foreign capital and remittances by non-resident Indians working in Gulf countries also inf luencing a positive relation of crude oil and stock market val- uation. Similarly, decrease in valuation of Indian currency against US dollar re- sulting to fall in market returns and higher import bills for crude. India being the net importing country to both crude oil and gold settles exchange transac- tions through US dollars. Hence, any changes in exchange rate of USD-INR have impacts on multiple economic indicators including stock market returns. This f luctuation in valuation of gold, exchange rate and market returns motivated to explore causal effect of relativity. India has two popular stock exchanges fa- cilitating to raise share capital and trade of listed securities. The sensitivity in- dex of Bombay stock exchange consisting of 30 most traded Indian companies’ forms indices known as Sensex. This paper shall be considering Sensex as a ba- rometer for understanding the Indian stock market returns as one of the vari- relAtivity of indiAn stock mArket… 103 able. The remaining part of this paper has been presented in following sections. Section 2 describes literature review, section 3 elucidates research methodol- ogy, section 4 presents results and discussions and finally section 5 concludes the research study with findings. Literature review This study reviewed scholastic research on gold, crude oil, exchange rate, and stock market returns from national and international sources of repute. The stock prices and exchange rates play a significant role in inf luencing investor’s confidence and development of the economy. Bhunia and Pakira (2014) and Ranjusha, Devasia and Nandakumar (2017) analysed the impact of exchange rate and gold on Sensex employing Granger causality and Johansen co-integra- tion test. The study finds existence of long run integration and no causal effect with exchange rate and gold. Geete (2016) assessed the effect of gold and dol- lar prices on stock exchange indices. The multiple regression analysis found positive correlation with gold and nifty whereas negative relationship between dollar and nifty. The relationship of gold on Karachi Stock Exchange (KSE) and Bombay Stock Exchange (BSE) were analysed by Bilal, Noraini Bt, Haq, Khan and Naveed (2013). The study found non-existence of long-run relationship amongst monthly average prices of gold and KSE stock index; whereas, a sig- nificant long-run relationship occurred between BSE stock index and average gold prices. Similarly, Narang and Singh (2012) argued absence of long run as- sociation and non-causal relation between prices of gold and Sensex. The de- pendence between global crude oil prices and stock indices in economies of fast emerging Asian countries were analysed by Mishra and Debasish (2019). The analysis revealed asymmetric effects with stock index returns and crude oil prices indicating better performance of Asian countries by means of higher production and consumption of goods and services. Studies of Sathyanarayana, Harish and Gargesha (2018); Najaf and Najaf (2016); Naifar and Al Dohaiman (2013) analysed volatility of crude oil prices and its inf luence on Indian stock market index. The statistical tests disclose that changes in crude prices have an impact on Sensex. It was also found that stock market index shares a direct relationship with crude oil, indicating raise in crude prices caused Sensex to go up and vice-versa. The short and long run asymmetric impact of oil prices, gold prices and their related volatilities on stock prices of emerging economies were A.N. Vijayakumar104 analysed by Raza, Syed, Aviral Kumar and Shahbaz (2016). The results indicat- ed non-linearity and absence of long run coefficients between the selected var- iables indicating higher volatility and decrease in stock prices. Similarly, Rah- man and Mustafa (2018) explored effect of gold and crude oil prices on US stock market variations and found there exists a long run convergence among the variables. The long and short-term impact on gold and stock returns were ana- lysed and found existence of co-integration relationship and non-existence of short-run relationship between the selected variables (Bhuyan & Dash, 2018). The interdependencies of oil, gold, exchange rate and stock market were ex- amined to identify linkages in Indian scenario by Mohanamani, Preethi and La- tha (2018). The study found negative linkage of crude oil prices and exchange rate with gold. It was also observed that exchange rate is highly inf luenced by changes in both oil and gold prices. Similarly, Sujit and Kumar (2011) found presence of co-integration amongst selected variables but Rastogi (2016) ar- gued there exists long term association. The relationship between oil prices, Nifty and BSE energy index were analysed and found absence of long run inte- gration between the variables by Sharma, Giri, Vardhan, Surange, Shetty and Shetty (2018). Bildirici and Turkmen (2015) analysed co-integration and caus- al relationship among oil and precious metals of gold, silver and copper. The statistical test revealed there exists a bi-directional relationship of oil, gold and silver. On the other hand Sari, Hammoudeh and Soyta (2010) argued that there exists weak long-run equilibrium relationship. The co-integration and non-lin- er causality amongst international gold, crude oil and Indian stock market were examined by Bouri, Jain, Biswal and Roubaud (2017) and found positive impact of implied volatility with the selected variables. Siddiqui and Seth (2015) as- sessed the relation of oil prices on market returns of India and found absence of integration and causality of oil prices with index. The co-integration relation- ships of gold price, crude oil price, exchange rates, Sensex and Nifty were ana- lysed by Bhunia (2013) and found they are closely interlinked with long term relationship. It was also argued that increase in crude oil prices lead to increase in production costs that will affect both stock prices and cash f lows. Interac- tive association amongst gold, crude oil prices and NT-US dollar exchange rate were examined and found these variables are remain considerably independ- ent from one another by Chang, Huang and Chin (2013). The dynamic interac- tions among oil, gold prices, exchange rate and the stock market in Indian con- text were analysed by Jain and Biswal (2016); Ingalhalli, Poornima and Reddy (2016). The empirical results found decline in crude oil and gold prices causes relAtivity of indiAn stock mArket… 105 fall in value of Indian currency. Ingalhalli, Poornima and Reddy (2016) initiate that there exists only unidirectional relationship among these variables. The studies on impact of macroeconomic elements on stock markets were also found during scholastic review. The inf luence of macroeconomic determi- nants such as Inf lation, rate of interest, exchange rate, industrial production index, money supply, silver, gold and crude oil prices on the Indian stock mar- ket indices were analysed by Venkatamuni Reddy, Nayak and Nagendra (2019) and Patel (2012). The statistical results revealed long run equilibrium relation and a causal relation between all macroeconomic factors and stock market in- dices. In addition, it was also found gold price, exchange rate and interest rate are positively correlated with four indices, however, crude oil price and silver price have positively correlated with three indices. Sekaran and Krishnamoor- thy (2016) analysed integration considering macro-economic variables on BSE index. The results revealed Chinese Yuan Renminbi, export of goods value, gold, inf lation, silver and US Dollar has inf luenced BSE sensex. Whereas, oil prices and Euro currency does not have a significant inf luence on BSE Sensex. The as- sociation between BSE Sensex with industrial production (IIP), inf lation, gold, interest rate, rate of exchange, foreign institutional investment (FII) and sup- ply of money were investigated by Mishra (2018) and found existence of caus- al relationship. Rakesh, Raju and Basavangowda (2016) studied the impact of currency f luctuations and found effect on Indian stock market as compared to Euro and Pound. The integration of stock prices and exchange rates were ex- amined and found co-integration and bidirectional relationship among stock price and exchange rates in Asia by Fauziah, Moeljadi and Ratnawati (2015). The presences of long memory in stock liquidity and returns of Indian equity market were examined by Bala and Gupta (2020) from the Indian context and confirmed presence of long memory in returns of all indices. Bidias-Menik and Tonmo (2020) tested predictive power of the implied forward rate of the term structure of interest rates at Africa. The study revealed that implicit forward rate does not have a significant predictive power in African countries. Volatility persistence of stock return in the market during pre and post meltdown were observed by Nageri (2019). This study considering return on the exchange dis- closed high volatility magnitude after the meltdown but low volatility magni- tude before the meltdown. The above scholastic evidences reviewed impact of crude oil, gold, interest rates, Index of Industrial Production (IIP), exchange rates, inf lation, Foreign Intuitional Investments (FII) and money supply on In- dian stock market. However, the findings are mystifying and there is a need for A.N. Vijayakumar106 clarity considering the recent economic developments on Indian market with a longer period of data. This dearth of scholastic literature motivated author to study relativity of stock market returns on exchange rate, gold and crude oil to add fresh insights to the existing domain of knowledge. Objective The objective is to determine relative impact of Indian currency exchange rate, gold and crude oil on Indian stock market in the short and long run. Research Methodology The study based on empirical data adopted causal research method with sec- ondary sources of monthly data for the period of 25 years from 1993 to 2018 considering 1200 observations. This study used Currency exchange rates (USD/ INR), Crude oil and Gold as independent variable and Indian stock market re- turns as dependant variable. The study collected secondary data of gold, crude oil and Sensex from web sources of New York Mercantile Exchange and Bombay Stock Exchange to ensure authenticity and accuracy. In the first step, stationar- ity of data using Augmented Dicky Fuller test has been checked. Subsequently, lag order selection criteria have been identified with the help of Akaike Infor- mation Criteria of data variables at level. Johansen co integration test is admin- istered to check the existence long term association of identified independent and dependent variables. The VAR model with VECM environment and Wald test is employed to understand long and short term relationship of independ- ent variables on market returns. After the satisfaction of results derived from the model, the study checked residual diagnostics. Augmented Dickey Fuller (ADF) test is used with the following regression equation: stock market. However, the findings are mystifying and there is a need for clarity considering the recent economic developments on Indian market with a longer period of data. This dearth of scholastic literature motivated author to study relativity of stock market returns on exchange rate, gold and crude oil to add fresh insights to the existing domain of knowledge. Objective The objective is to determine relative impact of Indian currency exchange rate, gold and crude oil on Indian stock market in the short and long run. Research Methodology The study based on empirical data adopted causal research method with secondary sources of monthly data for the period of 25 years from 1993 to 2018 considering 1200 observations. This study used Currency exchange rates (USD/INR), Crude oil and Gold as independent variable and Indian stock market returns as dependant variable. The study collected secondary data of gold, crude oil and Sensex from web sources of New York Mercantile Exchange and Bombay Stock Exchange to ensure authenticity and accuracy. In the first step, stationarity of data using Augmented Dicky Fuller test has been checked. Subsequently, lag order selection criteria have been identified with the help of Akaike Information Criteria of data variables at level. Johansen co integration test is administered to check the existence long term association of identified independent and dependent variables. The VAR model with VECM environment and Wald test is employed to understand long and short term relationship of independent variables on market returns. After the satisfaction of results derived from the model, the study checked residual diagnostics. Augmented Dickey Fuller (ADF) test is used with the following regression equation: ∆𝑦𝑦� = 𝑎𝑎 𝑎 𝑎𝑎𝑦𝑦��� 𝑎 ∑ 𝑏𝑏� ∆𝑦𝑦��� 𝑎 𝜀𝜀����� (1) ∆𝑦𝑦� = 𝑎𝑎 𝑎 𝑎𝑎𝑎𝑎 𝑎 𝑎𝑎𝑦𝑦��� 𝑎 ∑ 𝑏𝑏����� ∆𝑦𝑦��� 𝑎 𝜀𝜀� (2) The unit root in 𝑦𝑦� where ∆𝑦𝑦���is the lagged difference to accommodate serial correlation in the errors, 𝜀𝜀� . 𝑘𝑘is the appropriate lag length. (1) stock market. However, the findings are mystifying and there is a need for clarity considering the recent economic developments on Indian market with a longer period of data. This dearth of scholastic literature motivated author to study relativity of stock market returns on exchange rate, gold and crude oil to add fresh insights to the existing domain of knowledge. Objective The objective is to determine relative impact of Indian currency exchange rate, gold and crude oil on Indian stock market in the short and long run. Research Methodology The study based on empirical data adopted causal research method with secondary sources of monthly data for the period of 25 years from 1993 to 2018 considering 1200 observations. This study used Currency exchange rates (USD/INR), Crude oil and Gold as independent variable and Indian stock market returns as dependant variable. The study collected secondary data of gold, crude oil and Sensex from web sources of New York Mercantile Exchange and Bombay Stock Exchange to ensure authenticity and accuracy. In the first step, stationarity of data using Augmented Dicky Fuller test has been checked. Subsequently, lag order selection criteria have been identified with the help of Akaike Information Criteria of data variables at level. Johansen co integration test is administered to check the existence long term association of identified independent and dependent variables. The VAR model with VECM environment and Wald test is employed to understand long and short term relationship of independent variables on market returns. After the satisfaction of results derived from the model, the study checked residual diagnostics. Augmented Dickey Fuller (ADF) test is used with the following regression equation: ∆𝑦𝑦� = 𝑎𝑎 𝑎 𝑎𝑎𝑦𝑦��� 𝑎 ∑ 𝑏𝑏� ∆𝑦𝑦��� 𝑎 𝜀𝜀����� (1) ∆𝑦𝑦� = 𝑎𝑎 𝑎 𝑎𝑎𝑎𝑎 𝑎 𝑎𝑎𝑦𝑦��� 𝑎 ∑ 𝑏𝑏����� ∆𝑦𝑦��� 𝑎 𝜀𝜀� (2) The unit root in 𝑦𝑦� where ∆𝑦𝑦���is the lagged difference to accommodate serial correlation in the errors, 𝜀𝜀� . 𝑘𝑘is the appropriate lag length. (2) The unit root in yt where ∆yt–i is the lagged difference to accommodate serial correlation in the errors, εt.k is the appropriate lag length. relAtivity of indiAn stock mArket… 107 Johansen co-integration validated long-run relationship between the varia- bles at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equa- tion is as under: Johansen co-integration validated long-run relationship between the variables at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦��� = 𝛽𝛽𝑦𝑦��� (3) The corresponding VEC model is ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (4) ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = � ������ � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. 𝛔𝛔�� 𝛔𝛔�� ℎ(𝑧𝑧� � 𝛼𝛼) (8) (3) The corresponding VEC model is Johansen co-integration validated long-run relationship between the variables at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦��� = 𝛽𝛽𝑦𝑦��� (3) The corresponding VEC model is ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (4) ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = � ������ � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. 𝛔𝛔�� 𝛔𝛔�� ℎ(𝑧𝑧� � 𝛼𝛼) (8) (4) Johansen co-integration validated long-run relationship between the variables at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦��� = 𝛽𝛽𝑦𝑦��� (3) The corresponding VEC model is ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (4) ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = � ������ � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. 𝛔𝛔�� 𝛔𝛔�� ℎ(𝑧𝑧� � 𝛼𝛼) (8) (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: Johansen co-integration validated long-run relationship between the variables at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦��� = 𝛽𝛽𝑦𝑦��� (3) The corresponding VEC model is ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (4) ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = � ������ � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. 𝛔𝛔�� 𝛔𝛔�� ℎ(𝑧𝑧� � 𝛼𝛼) (8) (6) Where: Johansen co-integration validated long-run relationship between the variables at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦��� = 𝛽𝛽𝑦𝑦��� (3) The corresponding VEC model is ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (4) ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = � ������ � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. 𝛔𝛔�� 𝛔𝛔�� ℎ(𝑧𝑧� � 𝛼𝛼) (8) = Maximum Likelihood Estimator (MLE), Johansen co-integration validated long-run relationship between the variables at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦��� = 𝛽𝛽𝑦𝑦��� (3) The corresponding VEC model is ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (4) ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = � ������ � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. 𝛔𝛔�� 𝛔𝛔�� ℎ(𝑧𝑧� � 𝛼𝛼) (8) = expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine ex- istence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascer- tained by an auxiliary regression equation as: Johansen co-integration validated long-run relationship between the variables at level. Hence, VAR model with VECM environment have been employed to understand the casual relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦��� = 𝛽𝛽𝑦𝑦��� (3) The corresponding VEC model is ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (4) ∆𝑦𝑦��� = 𝛼𝛼��𝑦𝑦����� − 𝛽𝛽𝑦𝑦������ + 𝜖𝜖��� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = � ������ � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. 𝛔𝛔�� 𝛔𝛔�� ℎ(𝑧𝑧� � 𝛼𝛼) (8) (7) Where, ϵ indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of hetero- scedasticity in the developed model. A.N. Vijayakumar108 relationship. VECM model is a restricted VAR used with nonstationary series are known to be co-integrated. The co-integrating equation is as under: 𝑦𝑦�,� = 𝛽𝛽𝑦𝑦�,� (3) The corresponding VEC model is ∆𝑦𝑦�,� = 𝛼𝛼��𝑦𝑦�,��� − 𝛽𝛽𝑦𝑦�,���� + 𝜖𝜖�,� (4) ∆𝑦𝑦�,� = 𝛼𝛼��𝑦𝑦�,��� − 𝛽𝛽𝑦𝑦�,���� + 𝜖𝜖�,� (5) The coefficient αi measures speed of adjustment at i-th endogenous variable towards the equilibrium. The study using Wald test examined significance of explanatory components in a model. The Wald test statistic is as under: 𝑊𝑊� = �� ����� � ����(�� ) = 𝐼𝐼�(𝜃𝜃�)[𝜃𝜃� − 𝜃𝜃�]� (6) Where: 𝜃𝜃� = Maximum Likelihood Estimator (MLE), 𝐼𝐼�(𝜃𝜃�)= expected Fisher information (evaluated at the MLE). The study used Breusch - Godfrey Serial Correlation LM Test to determine existence of serial correlation of residuals under the developed model. This is an autocorrelation test used to check errors in the model. This statistics is ascertained by an auxiliary regression equation as- 𝑦𝑦� = 𝑋𝑋� 𝛽𝛽 + 𝜖𝜖� (7) Where, 𝜖𝜖 indicates errors. The study also used Breusch - Pagan - Godfrey test to check validity of heteroscedasticity in the developed model. σ�� ��� ℎ(𝑧𝑧� � 𝛼𝛼) (8) Where, 𝑧𝑧� indicates vector independent variable. This vector contains regressors from the original least square regression, it is tested by completing an auxiliary regression of the squared residual of the original equation on (1, 𝑧𝑧� ). Subsequently, Jarque Bera Test is administered for analysing normality of residuals. The test statistics measures difference of skewness and kurtosis of normal distribution series. It is computed as: 𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽 𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽 = �� ( 𝑆𝑆 � + (���) � � ) (9) (8) Where, zt indicates vector independent variable. This vector contains regres- sors from the original least square regression, it is tested by completing an aux- iliary regression of the squared residual of the original equation on (1, zt ). Subsequently, Jarque Bera Test is administered for analysing normality of residuals. The test statistics measures difference of skewness and kurtosis of normal distribution series. It is computed as: Where, 𝑧𝑧� indicates vector independent variable. This vector contains regressors from the original least square regression, it is tested by completing an auxiliary regression of the squared residual of the original equation on (1, 𝑧𝑧� ). Subsequently, Jarque Bera Test is administered for analysing normality of residuals. The test statistics measures difference of skewness and kurtosis of normal distribution series. It is computed as: 𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽 𝐽𝐽𝐽𝐽𝐽𝐽𝐽𝐽 𝐽 �� ( 𝑆𝑆 � + (���) � � ) (9) Finally, the stability of OLS model was tested using CUSUM test. This test is a cumulative sum portrayed at 5% critical line. Under this test, the plotted line should reside within the two measured lines otherwise the instability of test shall be noticed. Results and Discussions The study focussing on its intended objective found optimal lag order criteria for developing VAR model. Statistical outcome is shown in table 1. Table 1. Lag order selection criteria Lag Log L LR FPE AIC SC HQ 0 -5450.965 NA 1.98e+11 37.36278 37.41314 37.38295 1 -3647.696 3544.783 955232.3 25.12120 25.37304* 25.22208 2 -3609.701 73.64718 821696.4* 24.97056* 25.42385 25.15213* 3 -3601.386 15.89021 866241.2 25.02319 25.67796 25.28546 4 -3587.002 27.09363 876136.4 25.03426 25.89049 25.37723 5 -3570.893 29.90061 875893.2 25.03351 26.09121 25.45718 6 -3556.188 26.89188* 884303.7 25.04238 26.30155 25.54675 7 -3544.378 21.27464 910910.1 25.07108 26.53171 25.65615 8 -3533.288 19.67225 943248.5 25.10471 26.76681 25.77048 Note - * indicates selected lag order by criterion LR: LR test statistic sequential modified (each test at 5% level) FPE: indicates Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion Source: authors’ own calculations using E-views. The table no. 1 portrays, value under Akaike information criteria (AIC) of 24.97056, is lesser than other lag selection criteria. The study, therefore, selects AIC criteria of 2 lags. This selected 2 lags shall be used for co-integration test and at VAR model. (9) Finally, the stability of OLS model was tested using CUSUM test. This test is a cumulative sum portrayed at 5% critical line. Under this test, the plotted line should reside within the two measured lines otherwise the instability of test shall be noticed. Results and Discussions The study focussing on its intended objective found optimal lag order criteria for developing VAR model. Statistical outcome is shown in table 1. Table 1. Lag order selection criteria Lag Log L LR FPE AIC SC HQ 0 -5450.965 NA 1.98e+11 37.36278 37.41314 37.38295 1 -3647.696 3544.783 955232.3 25.12120 25.37304* 25.22208 2 -3609.701 73.64718 821696.4* 24.97056* 25.42385 25.15213* 3 -3601.386 15.89021 866241.2 25.02319 25.67796 25.28546 4 -3587.002 27.09363 876136.4 25.03426 25.89049 25.37723 5 -3570.893 29.90061 875893.2 25.03351 26.09121 25.45718 6 -3556.188 26.89188* 884303.7 25.04238 26.30155 25.54675 7 -3544.378 21.27464 910910.1 25.07108 26.53171 25.65615 8 -3533.288 19.67225 943248.5 25.10471 26.76681 25.77048 relAtivity of indiAn stock mArket… 109 Lag Log L LR FPE AIC SC HQ Note - * indicates selected lag order by criterion LR: LR test statistic sequential modified (each test at 5% level) FPE: indicates Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion S o u r c e : authors’ own calculations using E-views. The table 1 portrays, value under Akaike information criteria (AIC) of 24.97056, is lesser than other lag selection criteria. The study, therefore, selects AIC cri- teria of 2 lags. This selected 2 lags shall be used for co-integration test and at VAR model. Test of Stationarity The study checked stationarity of selected independent and dependant varia- bles and found results as shown under table 2. Table 2. Test of stationarity Sl. No Variable (log) At the level At first difference T-statistic P - value T - statistic P - value 1. Crude Oil -1.681562 0.4396 -8.347013 0.0000 2. Gold -0.631655 0.8600 -4.149046 0.0010 3. Exchange rate (USD/INR) 0.942455 0.7737 -4.004071 0.0016 4. Sensex -9.73271 0.6374 -10.31464 0.0000 S o u r c e : authors’ own calculations using E-views. The aforementioned table 2 depicts test of stationarity with T-statistics and P value at the level and first order. The P value of crude oil at the level 0.4396 whereas at the first difference, the variables became stationary showing P val- ue of 0.0000. Similarly, the value of variable Gold at level is 0.8600 whereas, at first difference it became 0.0010 indicating stationarity. The exchange rate Table 1. Lag… A.N. Vijayakumar110 variable converted to stationarity at the first difference with the P value of 0.0016. From the above process, variables became stationary at the first differ- ence were found to be satisfactory and essential for further analysis. Test of Co-integration The study used prices of oil, gold, exchange rate and monthly returns from one of the Indian exchange indices such as Sensex of Bombay Stock Exchange. The table 3 indicates statistical output of Johansen co-integration test. Table 3. Unrestricted Co integration Test (trace) Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0.05 Critical Value Prob.** None * 0.233395 103.6983 47.85613 0.0000 At most 1 0.047716 24.76069 29.79707 0.1702 At most 2 0.033215 10.23972 15.49471 0.2627 At most 3 0.000698 0.207266 3.841466 0.6489 Trace test indicates 1 cointegrating at the 0.05 level * rejection of hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values S o u r c e : authors’ own calculations using E-views. The aforementioned table 3 indicates that the P value is less than 0.05 at 95% confidence level advising for rejecting null hypothesis of no co-integration of independent variables with dependant variable (Indian stock market returns). The study, therefore, accepts alternative hypothesis of existence of co-integra- tion of independent variables with dependant variable having a P value more than 0.05. This indicates long-run association amongst the variables of crude oil, gold, exchange rate and the Indian stock market. relAtivity of indiAn stock mArket… 111 Vector Error Correction Model The study after finding the existence of co-integration of independent variables with dependant variables administered Vector Error Correction Model using 2 lags, as per AIC lag order selection criteria. Accordingly, the following equation has been obtained with the developed Error Correction model: * rejection of hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values Source: authors’ own calculations using E-views. The aforementioned table 3 indicates that the P value is less than 0.05 at 95% confidence level advising for rejecting null hypothesis of no co-integration of independent variables with dependant variable (Indian stock market returns). The study, therefore, accepts alternative hypothesis of existence of co-integration of independent variables with dependant variable having a P value more than 0.05. This indicates long-run association amongst the variables of crude oil, gold, exchange rate and the Indian stock market. Vector Error Correction Model The study after finding the existence of co-integration of independent variables with dependant variables administered Vector Error Correction Model using 2 lags, as per AIC lag order selection criteria. Accordingly, the following equation has been obtained with the developed Error Correction model – 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷7𝐷𝐷 𝐷 𝐷𝐷𝐶𝐶 𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐶7𝐷7𝐶𝐷𝐶𝐷𝐶𝐶𝐷𝐷 𝐷 𝐶𝐶𝐶𝐶𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐶7𝐶𝐶𝐶𝐷𝐶𝐷𝐷𝐶𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐶𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐶𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐶𝐶𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷7𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷𝐶𝐶𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐶𝐷 𝐷 𝐷𝐷𝐷𝐶𝐶𝐶𝐶𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷𝐷𝐶𝐶𝐶𝐶𝐷𝐷𝐷𝐷𝐷 𝐷 𝐷𝐷 The study using above equation with least squares method estimated the VAR model as under. Table 4. Model output Dependent Variable: D(RSNSX) Method: Least Squares Sample (adjusted): 1993M07 2018M03 Included observations: 297 after adjustments D(RSNSX) = C(1)*( RSNSX(-1) + 0.0232049340022*USIN(-1) + 0.000921902212722*CL(-1) - 0.000579761608521*GL(-1) - 1.87566064959 ) + C(2)*D(RSNSX(-1)) + C(3)*D(RSNSX(-2)) + C(4) *D(USIN(-1)) + C(5)*D(USIN(-2)) + C(6)*D(CL(-1)) + C(7)*D(CL(-2)) + C(8)*D(GL(-1)) + C(9)*D(GL(-2)) + C(10) Coefficient Std. Error t-Statistic Prob. (10) The study using above equation with least squares method estimated the VAR model as under. Table 4. Model output Dependent Variable: D(RSNSX) Method: Least Squares Sample (adjusted): 1993M07 2018M03 Included observations: 297 after adjustments D(RSNSX) = C(1)*( RSNSX(-1) + 0.0232049340022*USIN(-1) + 0.000921902212722*CL(-1) - 0.000579761608521*GL(-1) - 1.87566064959 ) + C(2)*D(RSNSX(-1)) + C(3)*D(RSNSX(-2)) + C(4) *D(USIN(-1)) + C(5)*D(USIN(-2)) + C(6)*D(CL(-1)) + C(7)*D(CL(-2)) + C(8)*D(GL(-1)) + C(9)*D(GL(-2)) + C(10) Coefficient Std. Error t-Statistic Prob. C(1) -0.878069 0.102668 -8.552475 0.0000 C(2) -0.118640 0.086014 -1.379313 0.1689 C(3) -0.011344 0.060229 -0.188340 0.8507 C(4) -0.677760 0.464040 -1.460564 0.1452 C(5) 0.265412 0.471521 0.562885 0.5740 C(6) 0.020208 0.096190 0.210087 0.8337 A.N. Vijayakumar112 Coefficient Std. Error t-Statistic Prob. C(7) 0.041096 0.095956 0.428280 0.6688 C(8) 0.002758 0.012718 0.216878 0.8285 C(9) 0.005024 0.012541 0.400599 0.6890 C(10) -0.001986 0.418485 -0.004746 0.9962 R-squared 0.490506 Mean dependent var -0.014913 Adjusted R-squared 0.474529 S.D. dependent var 9.627543 S.E. of regression 6.978945 Akaike info criterion 6.756763 Sum squared resid 13978.53 Schwarz criterion 6.881131 Log likelihood -993.3793 Hannan-Quinn criter. 6.806552 F-statistic 30.70049 Durbin-Watson stat 2.000781 Prob(F-statistic) 0.000000 S o u r c e : authors’ own calculations using E-views. The table 4 depicts co-efficient of C1 is negative (–0.878069) and significant with P value 0.0000, i.e., less than 0.05. This indicates presence of long run re- lationship from exchange rate (USD/INR), crude oil and gold to Indian stock market returns. It is observed that F – statistics (30.70049) is significant with P value less than 0.05 (0.000000) showing combined inf luence of independent variables on dependant variable. The value of F statistics is more important than R square equivalent to 50% according to the model. Short Run Causal Relationship The study using Wald test explored short run causal relationship of exchange rate, crude oil, gold and Indian stock market. Accordingly, the table 5 depicts the Chi-square value with corresponding P values. The statistical results can- not reject null Hypothesis of no short-run causality and concluded the non-ex- istence of short run causal relationship of exchange rate (USD/INR), crude oil and gold with Indian stock market. Table 4. Model… relAtivity of indiAn stock mArket… 113 Table 5. Wald Test Variables Chi-Square value P-value Exchange Rate 2.441209 0.2951 Crude Oil 0.336526 0.8451 Gold 0.228502 0.8920 Restrictions are linear in coefficients. S o u r c e : authors’ own calculations using E-views. Serial Correlation Test The study checked validity of OLS model through residual diagnostic test of Breusch-Godfrey Serial Correlation. Accordingly, the observed R square (0.291429) with corresponding probability value (0.8644) is more than 5%. Hence, the null hypothesis - Residuals are not serially correlated, cannot be re- jected. The output of results is desirable. Test of Heteroskedasticity The study finds rejection of null hypothesis from the statistical value of ob- served R square (31.03625) with corresponding probability value (0.0019) be- ing less than 0.05 hence, it finds that residuals are homoscedastic. Test of Normality Jarque Bera Normality test finds the p value (0.103) is not significant at the gen- eral acceptance level of 5% (figure 1). Therefore, the study cannot reject null hypothesis, and concludes that residuals are being normally distributed. It is also fulfilling the condition of bell-shaped curve. A.N. Vijayakumar114 Figure 1. Test of Normality Exchange Rate 2.441209 0.2951 Crude Oil 0.336526 0.8451 Gold 0.228502 0.8920 Restrictions are linear in coefficients. Source: authors’ own calculations using E-views. Serial Correlation Test The study checked validity of OLS model through residual diagnostic test of Breusch-Godfrey Serial Correlation. Accordingly, the observed R square (0.291429) with corresponding probability value (0.8644) is more than 5%. Hence, the null hypothesis - Residuals are not serially correlated, cannot be rejected. The output of results is desirable. Test of Heteroskedasticity The study finds rejection of null hypothesis from the statistical value of observed R square (31.03625) with corresponding probability value (0.0019) being less than 0.05 hence, it finds that residuals are homoscedastic. Test of Normality Jarque Bera Normality test finds the p value (0.103) is not significant at the general acceptance level of 5% (figure 1). Therefore, the study cannot reject null hypothesis, and concludes that residuals are being normally distributed. It is also fulfilling the condition of bell-shaped curve. Figure 1. Test of Normality Source: authors’ own calculations using E-views. S o u r c e : authors’ own calculations using E-views. Stability Test Stability of VECM model developed in the study is subjected to CUSUM test. Figure 2 shows that curved line is between +15 and -15. Hence, the developed VECM model under the study is stable and it is desirable to accept the results. Figure 2. Stability test Stability Test Stability of VECM model developed in the study is subjected to CUSUM test. Figure 2 shows that curved line is between +15 and -15. Hence, the developed VECM model under the study is stable and it is desirable to accept the results. Figure 2. Stability test Source: authors’ own calculations using E-views. Findings and Conclusion Stock market responds to the present and subsequent economic developments and shocks. The potential of earnings by listed companies shall reflect in the index movements. The general economic theory of cost and expected returns attracts investors to divert there investments on potential assets. This study considering shocks of exchange rate, gold and oil on Indian stock market evidenced long-run association of co-movements with the help of Johansen co- integration test. The Vector auto regression with vector error correction model finds long run causal relationship of identified variables on Indian market, Sensex. However, the study did not find the existence of short run causal relationship of exchange rate, oil and gold with Sensex. This research paper validated with economic and investment theory of influencing prices of crude oil, gold and exchange rate of stock market. Thus, increase in crude oil prices and exchange rate (USD/INR) would lead to plunge in returns at Indian stock market in the long run. Similarly, increase in import of gold leads to weakening of Indian rupee and results in declining market returns. This findings of study may be useful to investor and fund managers to develop appropriate strategies for manging the port-folio and investment decisions. Further research studies may consider daily or weekly data of gold, crude oil and S o u r c e : authors’ own calculations using E-views. relAtivity of indiAn stock mArket… 115  Findings and Conclusion Stock market responds to the present and subsequent economic developments and shocks. The potential of earnings by listed companies shall ref lect in the index movements. The general economic theory of cost and expected returns attracts investors to divert there investments on potential assets. This study considering shocks of exchange rate, gold and oil on Indian stock market evi- denced long-run association of co-movements with the help of Johansen co-in- tegration test. The Vector auto regression with vector error correction mod- el finds long run causal relationship of identified variables on Indian market, Sensex. However, the study did not find the existence of short run causal rela- tionship of exchange rate, oil and gold with Sensex. This research paper validat- ed with economic and investment theory of inf luencing prices of crude oil, gold and exchange rate of stock market. Thus, increase in crude oil prices and ex- change rate (USD/INR) would lead to plunge in returns at Indian stock market in the long run. Similarly, increase in import of gold leads to weakening of In- dian rupee and results in declining market returns. This findings of study may be useful to investor and fund managers to develop appropriate strategies for manging the port-folio and investment decisions. Further research studies may consider daily or weekly data of gold, crude oil and exchange rates to estimate impact on market returns and also may compare with developing countries of economical factors to identify opportunities for wealth creation at Indian and other identified stock markets. Acknowledgement “The author is grateful to editor and anonymous reviewers of the journal for their extremely useful suggestions to improve the quality of the paper. Usual disclaimers apply.” Declaration of Conf licting Interests The authors declared no potential conf licts of interest with respect to the re- search, authorship and/or publication of this article. A.N. Vijayakumar116  References Bala, A., & Gupta, K. (2020). Examining the Long Memory in Stock Returns and Liquid- ity in India. Copernican Journal of Finance & Accounting, 9(3), 25-43. http://dx.doi. org/10.12775/CJFA.2020.010. Bhunia, A. (2013). Cointegration and Causal Relationship among Crude Price, Domestic Gold Price and Financial Variables: An Evidence of BSE and NSE. Journal of Contem- porary Issues in Business Research, 2(1), 1-10. Bhunia, A., & Pakira, S. (2014). Investigating the Impact of Gold Price and Exchange Rates on Sensex: An Evidence of India. European Journal of Accounting, Finance and Business, 2(1), 1-11. Bhuyan, A.K., & Dash, A.K. (2018). A dynamic causality analysis between gold price movements and stock market returns: Evidence from India. Journal of Manage- ment Research and Analysis, 5(2), 117-124. http://dx.doi.org/10.18231/2394- 2770.2018.0019. Bidias-Menik, H.P., & Tonmo, S.G. (2020). Interest rate predictability in some selected African countries. Copernican Journal of Finance & Accounting, 9(3), 45-60. http:// dx.doi.org/10.12775/CJFA.2020.011. Bilal, A.R., Noraini Bt, A.T., Haq, I.U., Khan, M.N.A.A, & Naveed, M. (2013). How Gold Pric- es Correspond to Stock Index: A Comparative Analysis of Karachi Stock Exchange and Bombay Stock Exchange. World Applied Sciences Journal, 21(4), 485-491. http:// dx.doi.org/10.5829/idosi.wasj.2013.21.4.2870. Bildirici, M.E., & Turkmen, C. (2015). Nonlinear Causality between Oil and Precious Metals. Resources Policy, 46(2), 202-211. http://dx.doi.org/10.1016/j.resour- pol.2015.09.002. Bouri, E., Jain, Biswal, P.C., & Roubaud, D. (2017). Cointegration and Nonlinear Causal- ity amongst Gold, Oil, and the Indian Stock Market: Evidence from Implied Vola- tility Indices. Resources Policy, 52, 201-206. http://dx.doi.org/10.1016/j.resour- pol.2017.03.003. Chang, H.-F., Huang, L.-C., & Chin, M.-C. (2013). Interactive Relationships between Crude Oil Prices, Gold Prices, and the NT-US Dollar Exchange Rate-A Taiwan Study. Energy Policy, 63, 441-448. http://dx.doi.org/10.1016/j.enpol.2013.09.029. Dalei, N.N., Roy, H., & Gupta, A. (2017). Crude Oil Import of India from Its Major Oil Trade Partner Countries: An Empirical Evidence Using Panel Data Analysis. International Journal of Advanced Research and Development 726(6), 2455-4030. Fauziah, F., Moeljadi, M., & Ratnawati, K. (2015). Dynamic relationship between ex- change rates and stock prices in Asia, 2009-2013. Journal of Economics, Finance and Accounting, 2(1), 124-124. http://dx.doi.org/10.17261/Pressacademia.201519783. Geete, V. (2016). A Study on Impact of Gold Prices and Dollar Prices on Stock Market In- dices with Special Reference to Sensex Nifty. European Journal of Economic and Busi- ness, 1(1), 28-33. http://dx.doi.org/10.20936/EJEB/160105. relAtivity of indiAn stock mArket… 117 Ingalhalli, V., Poornima B.G., & Reddy, Y.V. (2016). A Study on Dynamic Relationship Be- tween Oil, Gold, Forex and Stock Markets in Indian Context. Paradigm, 20(1), 83-91. http://dx.doi.org/10.1177/0971890716637706. Jain, A., & Biswal, P.C. (2016). Dynamic Linkages among Oil Price, Gold Price, Ex- change Rate, and Stock Market in India. Resources Policy, 49, 179-185. http://dx.doi. org/10.1016/j.resourpol.2016.06.001. Mishra, P. (2018). An Investigation of the Macroeconomic Factors Affecting the Indi- an Stock Market. Australasian Accounting, Business and Finance Journal, 12(2),71-86. http://dx.doi.org/10.14453/aabfj.v12i2.5. Mishra, S., & Debasish, S.S. (2019). The Quantile Dependence between Global Crude Oil Price and Stock Markets in Emerging Asia: Evidence from Major Oil Consuming Nations. Afro-Asian Journal of Finance and Accounting, 9(3), 309-331. http://dx.doi. org/10.1504/AAJFA.2019.100980. Mohanamani, P., Preethi, D., & Latha, L. (2018). Dynamic Linkage between Gold, Oil, Ex- change Rate and Stock Market Returns: Evidence from India. International Journal of Pure and Applied Mathematics, 119(17), 2567-2580. Nageri, K.I. (2019). Evaluating volatility persistence of stock return in the pre and post 2008-2009 financial meltdown. Copernican Journal of Finance & Accounting, 8(3), 75-94. http://dx.doi.org/10.12775/CJFA.2019.013. Naifar, N., & Al Dohaiman, M.S. (2013). Nonlinear Analysis among Crude Oil Prices, Stock Markets’ Return and Macroeconomic Variables. International Review of Eco- nomics and Finance, 27, 416-431. http://dx.doi.org/10.1016/j.iref.2013.01.001. Najaf, R., & Najaf, K. (2016). Impact of Crude Oil Prices on the Bombay Stock Exchange. Journal of Business & Financial Affairs, 5(4), 10-12. http://dx.doi.org/10.4172/2167- 0234.1000216. Narang, S.P., & Singh, R.P. (2012). Causal Relationship between Gold Price and Sensex: A Study in Indian Context. Vivekananda Journal of Research, 1, 33-37. Patel, S. (2012). The effect of Macroeconomic Determinants on the Performance of the Indian Stock Market. NMIMS Management Review, 22. Rahman, M., & Mustafa, M. (2018). Effects of Crude Oil and Gold Prices on US Stock Market: Evidence for USA from ARDL Bounds Testin. Finance and Market, 3(1), 1-9. http://dx.doi.org/10.18686/fm.v3.1055. Rakesh, D., Raju, J.K., & Basavangowda, K.G. (2016). An Impact of Currency Fluctuations on Indian Stock Market. International Journal of Application or Innovation in Engi- neering & Management, 5(6), 146-151. Ranjusha, N., Devasia, M.D., & Nandakumar, V.T. (2017). Cointegrating Relation Be- tween Exchange Rate and Gold Price. International Journal of Research-Granthaal- ayah, 5(10), 263-269. http://dx.doi.org/10.5281/zenodo.1043230. Rastogi, S. (2016). Gold Price, Crude Oil, Exchange Rate and Stock Markets. International Journal of Corporate Finance and Accounting, 3(2), 1-13. http://dx.doi.org/10.4018/ IJCFA.2016070101. Raza, N., Syed, J.H.S., Aviral Kumar, T., & Shahbaz, M. (2016). Asymmetric Impact of Gold, Oil Prices and Their Volatilities on Stock Prices of Emerging Markets. Resourc- es Policy, 49, 290-301. A.N. Vijayakumar118 Sari, R., Hammoudeh, S., & Soyta, U. (2010). Dynamics of Oil Price, Precious Metal Prices, and Exchange Rate. Energy Economics, 32(2), 351-362. http://dx.doi.org/10.1016/j. eneco.2009.08.010. Sathyanarayana, S., Harish, S.N., & Gargesha, S. (2018). Volatility in Crude Oil Prices and its Impact on Indian Stock Market Evidence from BSE Sensex. SDMIMD Journal of Management, 9(1), 65-78. http://dx.doi.org/10.18311/sdmimd/2018/19997. Sekaran, M.C., & Krishnamoorthy, V. (2016). Inf luence of selected Macro Economics variables on Indian stock market – A study of BSE Sensex. Abhinav National Monthly Refereed Journal of Research in Commerce & Management, 5(3). Sharma, A., Giri, S., Vardhan, H., Surange, S., Shetty, R., & Shetty, V. (2018). Relationship between Crude Oil Prices and Stock Market: Evidence from India. International Jour- nal of Energy Economics and Policy, 8(4), 331-337. Siddiqui, S., & Seth, N. (2015). Do Global Oil Price changes affect Indian Stock Market Re- turns? Journal of Management and Public Policy, 6(2), 29-41. Sujit, K.S., & Kumar, B.R. (2011). Study on Dynamic Relationship among Gold Price, Oil Price, Exchange Rate and Stock Market Returns. International Journal of Applied Business and Economic Research, 9(2), 145-165. Venkatamuni Reddy, A.R., Nayak, R., & Nagendra, S. (2019). Impact of Macro - Eco- nomic Factors on Indian Stock Market- A Research of BSE Sectoral Indices. Inter- national Journal of Recent Technology and Engineering, 8(2), 597-602. http://dx.doi. org/10.35940/ijrte.B1110.0782S719.