THE EFFECT OF OIL PRICE ON STOCK MARKET RETURNS WITH MODERATING EFFECT OF FOREIGN DIRECT INVESTMENT & FOREIGN PORTFOLIO INVESTMENT: EVIDENCE FROM PAKISTAN STOCK MARKET
^{1}Reasearch Scholar, Karachi University Business School, University of Karachi, Pakistan
^{2}Associate Professor, Karachi University Business School, University of Karachi, Pakistan.
ABSTRACT
This paper investigates the moderating impact of FDI & FPI in the association of macroeconomic variables along with Oil prices & Index returns. Monthly data has been used from the period 2005 to 2018. Efficient unit root & breakpoint unit root tests results indicate that all variables are stationary at 1st difference. Cointegration test results signify the presence of longrun relationship in model. GARCH (1,1) model has been applied for analyzing the volatility in the data series. Furthermore, least square method is employed to check dependency & fitness level of model. In order to investigate the moderating impact, regression technique has been applied. Findings of LSM technique indicate that index returns aren’t significantly dependent on macroeconomic variables on 1st difference of data series because variables predicting behavior has been changed with respect to stationarity of data. Exchange rate & interest rate have negative significant association with index returns. Oil prices & foreign direct investment have positive relationship with stock market return. FDI & FPI are unable to moderate significantly model dynamics. For estimating the panel regression model, 11 different sectors data is used and results show that exchange rate & oil prices have positive significant impact on sector wise price change but interest rate has significant negative association.
Keywords:Exchange rate, Interest rate, Foreign direct & portfolio , investment, Stock market return, Oil prices, Cointegration, Stationarity, GARCH (1,1).
JEL Classification:C510; E440; G110.
ARTICLE HISTORY: Received: 4 December 2018, Revised: 18 January 2019, Accepted: 26 February 2019, Published: 11 April 2019
Contribution/ Originality: This study contributes in the existing literature of finance through checking moderating impact of FDI & FPI in the relation of macroeconomic variables along with oil prices from KSE 100 index return. Moreover, findings of study allow investors & government to take better decision regarding investment philosophy in volatile market.
Stock market volatility, due to macroeconomic risk factors & less investor confidence is a fundamental concept in finance literature & many economist highlighted this issue, as Sadorsky (2001) documented that association between oil price shocks, macroeconomic factors & financial variables have a dynamic process. From decades, scholars tried to discover the responsiveness of equity market with respect to risk factors of macroeconomic variables & from last few years, this concept get an enormous attention of economists. For diversifying risk, developing understanding about the complex link between macroeconomic variables & index fluctuations is crucial (Khan et al., 2018). Oil price & stock market return have a dynamic association, which should be explore in detail.
First methodology had been developed on that model by Ewing and Thompson (2007). Lee and Chiou (2011) Imply a univariate method for investigating the relation across oil prices & index return. In 1980, developed countries reduced the restrictions on investing in financial markets by financial liberalization concept & due to that concept foreign investment theory emerged. Portfolio investment & direct investment are the two main components of foreign investment in any market. FPI has a temporary affect whereas FDI has a long term influence on the economic indicators (Lipsey et al., 1999). Ferrer et al. (2016) argues that high interest rates attracts the foreign portfolio investment. Therefore, emerging economies always need to get foreign investments in both ways either direct or portfolio (Broto et al., 2011).
As per current data released by Ministry of Finance, petroleum sector import bill is 24% of total import bill. Asian countries are the major consumers of crude oil although falling demand in Asian region lead to fall in the global demand of Oil of any country (Sadorsky, 2001) however, oil prices are the most fundamental driver in the financial growth of country’s economy in terms of inflow of money for oil producing countries & expense for oil consumption countries. Hence energy sector is the major contributor (Hamilton, 1983).
In developing countries, stock market return depends on the economic indicators. The performance of PSX index from last two years was flourishing, Filis et al. (2011) although massive decline had observed within the last quarter of this year which indicate high volatility in the market, uncertain situation made investors unable to clearly get the risk factors behind that (Khan, 2017).
Oil prices changes are affected by the different macroeconomic channels of a country. Oil importing countries like Pakistan recorded decline in balance of payment, which subsequently put downward pressure on exchange rates, making imports more costly and exports less profitable consequently real national income fall. This is because of extensive oil usage for transportation, power, and input for production process. Therefore, effective decisions require this information.
Pakistan imports large quantity of crude oil on annual basis. Furthermore, country’s population is growing significantly faster than the availability of oil & its relevant items for household, textile, agribusiness & others.
As stock market is a major factor for measuring economic growth & in case of Pakistan where many development projects are in process, stock market behavior is the fundamental aspect which needs to be explored. Currently, due to CPEC country is facing an energy crises or trade deficit problems. Whereas, import bill consist of 25% weighted of oil import. So in a whole scenario stock market behavior with respect to macroeconomic variables is a phenomena which needs to be addressed for effective decision making.
This study also helps to develop understanding about how macroeconomic elements influence share price of corporations & different market sectors. This is also an essential prerequisite for investors, portfolio managers, analysts, economists & policy makers to make better investing decisions in a current scenario of country’s economy because from last few years stock market index drops significantly. However, market faced an anonymous challenges, which needs to be addressed for developing an appropriate understanding about the market fluctuations (currency devaluation, political instability, security threats etc.) (Ferrer et al., 2016). Moreover, it help to get an insight about the implications of investment philosophy in a diverse domain of financial management such as, portfolio management, asset allocation, mitigating risk, fixed income analysis & fiscal or monetary policy planning.
This exploration is an effort to fill the gap between literature & financial market of Pakistan through exploring moderating effect of foreign direct investment & foreign portfolio investment in the relation of exchange rate fluctuations, oil price volatility & interest rate fluctuations from stock market returns moreover, also perform sectoral basis analysis on eleven different sector along with same determinants as well. There is no evidence is available from previous literature regarding moderating impact of FDI & FPI on key economic variables of economy. Moreover, no previous research has been conducted in the context of Pakistan financial market along with that model specification.
In recent years, Pakistan as an emerging economy encountered from many fundamental risk factors i.e. (interest rate, exchange rate & Oil Prices) moreover, it has been observed that equity market is unable to respond all that financial crises. Monetary policy statement of SBP reported that Pakistan foreign reserves declined to its historical level of $9 Billion & GDP growth is reduced to 5% annually. On the other hand monthly current account deficit has been inclined to $3 billion, which is also effected by the continuous decline in exports. Therefore the current scenario of Pakistan’s economy contains various risk factors which leads towards the loss of investor’s confidence. According to Sadorsky (2001) risk & returns of equity are the major tradeoffs of each other & uncertain economic condition of a country’s economy may the cause of losing investors’ confidence (MehrUnNisa and Nishat, 2011) when investor were unaware regarding which risk factors has been driven the market so, investment become more reluctant.
The inconsistency of Pakistan’s equity market has encountered a massive shortage of investments & because of that stock market is unable to properly cater the FPI in daily trading. On the other hand stock market performance declines when there is a decline in exchange rates due to the increase in Oil Price (Sadorsky, 2001). Rahman and Mustafa (2018) suggested in a long run Oil prices has very insignificant effect on stock returns but Phan et al. (2018) explore that Crude Oil price uncertainty has a negative significant effect on corporate investment. There are different scholars that have various perspectives with regards of association across macroeconomic variables & stock market deviations, Sathyanarayana et al. (2017) oil price shocks has a huge potential to transmit fluctuations in to stock return, but in case of oil producing countries, oil prices leads to high returns (Sadorsky, 2001) when development project set aside in a country hence macroeconomic variables react differently.
In a current scenario of Pakistan, mega developmental projects is in process & our foreign reserves are also been decreases on continuous basis & due to that foreign investor losses their confidence. Previous literature never explored this phenomena as well as none of the previous researcher focused on the moderating impact of foreign portfolio investment and its relationship with exchange rate, interest rate, & oil prices with respect to various industry’s stock returns.
This research is an exertion to fill the gap between literature & financial market of Pakistan through exploring the moderating effect of foreign direct investment in the relation of exchange rate fluctuations, oil price volatility & interest rate fluctuations from stock market returns on sectoral basis as well as no insight has been given from previous literature regarding moderating impact of FDI on key economic variables of economy, because of that practitioners in market are not quite familiar with moderating impact of FDI in the context of Pakistan investment market. Moreover, no previous research has been conducted in the context of Pakistan financial market with that variables or model specification.
This study effort to explore empirically:
For developing testable hypothesis discussing the dimensions of price uncertainty of crude oil & how it’ll effect on share price of different companies because when prices of crude oil fluctuated so due to that inflationary pressures also adjusted in the economy & in response of that currency & interest rate fluctuations occur. Whereas, country’s economy is highly vulnerable on foreign sources & stock market is one of the fundamental determinant of economy & from last year stock market faces a number of challenges such as political instability, circular debt, high trade deficit, low growth in GDP & continuous decline in reserves. All that elements leads to high risk in a market & behavior of foreign investor changes, as portfolio theory by Sadorsky (2001) documented that higher risk leads to high risk premium which causes the change in investment decision. So, in a whole scenario exploring the moderating relationship across model is an extensive need of a literature & moreover, it hasn’t been addressed previously.
Ha1: There is a significant relationship between macroeconomic variables & kse 100 index return.
Ha2: There is a significant relationship between macroeconomic variables & kse 100 index sectoral basis return.
Ha3: Moderating impact of foreign direct investment in the relationship of macroeconomic variables & kse 100 index return.
Ha4: Moderating impact of foreign portfolio investment in the relationship of macroeconomic variables & kse 100 index return.
There are several macroeconomics factors that affect the returns on the stocks of different sectors of economy. Many researchers has worked on this phenomena by taking different sectors and variables. According to MehrUnNisa and Nishat (2011) explore the preceding behavior of stock prices, company size, previous earnings per share are the most imperative elements. However, macroeconomic factors like, GDP growth, rate of interest and financial debt has strong association with index fluctuations. Market to book value, share turnover ratio and inflation can also effect the index deviations. While prices of crude oil also impact significantly on stocks returns. As Sadorsky (2001) found that the increased prices of crude oil effect positively on the return of Canadian oil and gas companies but term premium and exchange rate effected negatively Whereas, when it comes to oil supply shocks on listed oil and gas companies stock returns it was examined by Ewing et al. (2018) that the declining supply of oil will effect on stock market differently (Gay, 2016) revealed that emerging market stocks doesn’t show any significant response to respective exchange rate or oil price variations & show the week foam of market efficiency (Maghyereh, 2004) in emerging economies do not have strong association among stock returns & oil price shocks as well no fluctuations occur due to variation of stock market (Aloui et al., 2012) in case of emerging stock markets oil price risk is relatively high moreover, during rising oil markets relationship become more significant (Cong et al., 2008). Oil price shocks don't have a significant impact on Chinese equity market except in manufacturing/oil & gas sector however, fluctuations among oil prices leads to speculation across mining & pharma companies & raise their stocks. Sadorsky (2001) oil demand fluctuations effect significantly on oil producing countries (Oyerinde, 2019). Found, foreign portfolio investment has long term significant association with stock market return moreover also suggested that exchange rate & inflation are the key indicators for deciding the trend of stock market especially in developing countries. While in Asian crises (Omay and Iren, 2019) foreign investors investment is more sensitive as their herding behavior while crises & due to that stock market become more volatile when market is concentrated towards foreign portfolio (Tsagkanos et al., 2018). Explored foreign direct investment has week symmetric influence on stock market.
Phan et al. (2018) Concluded that price uncertainty of crude oil has an inverse effect on corporate investment however, economies that are oil producers effected more as compared to oil consuming economies (Rahman and Mustafa, 2018). Moreover, significant change for gold prices but insignificant for Oil but in long run, relationship become change from negative to equilibrium point (Narayan and Narayan, 2010). Foreign portfolio investment held in the same period & second is change in number of participants in the stock market. Similarly, Sathyanarayana et al. (2017) also found the significant positive impact of crude oil price volatility on Indian stock market or even have a huge potential or competency to transmit shocks over it.
Faff and Brailsford (1999) Results reveal that positive oil price sensitivity in the Oil and Gas and Diversified Resources industries. While negative oil price sensitivity in the Paper and Packaging, and Transport industries. Waheed et al. (2018) found direct strong association between oil price changes & stock return (Filis et al., 2011) found strong negative relationship between stock market prices and oil price in case of any noneconomic crisis.
Tahmoorespour et al. (2018) showcased that price volatility of oil, mainly impact the oil & gas industry as well as chronological impact on mining industry but have an insignificant or least effected food & beverage industry. Arouri (2011) findings, highly significant linkage across Oil Prices & Stock market returns in majority of the sectors of European market but sensitivity of market against Oil prices dependent on the characteristics of the sectors (Ayub, 2018) ﬁndings indicate that the volatility is transmitted to the returns of exchange rate and diﬀerent commodities & the volatility spillover is also observed from crude oil to exchange rate, equity market, sugar and palm oil.
Khan et al. (2018) found no causal link between exchange rate & stock prices (Rehman et al., 2018) indicated that India, Bangladesh & Pakistan has not be the random walk process markets & foam inefficient hypothesis means all three markets are more predictable due to consistency in the returns While discussion of Javorci (2004) study revolved around a very famous ideology that foreign direct investments (FDI) are always beneficial for any growing economy and observed a consistent positive impact due to the presence of FDI. A one standarddeviation increase in the foreign presence in downstream sectors is results in a 15percent rise in output of each domestic ﬁrm in supplying industries.
Oil price shocks on stock market also vary with respect to longrun & short run period. Oil price variations effect is dynamic on stock market (sectorwise), as food/beverage sector has a lesser effect than the financial & energy sector which is largely dependent on oil prices (Tahmoorespour et al., 2018). Chen et al. (1987) proposed the multifactor asset pricing model which is used to predict the direction of macroeconomic variables in the economy & also helpful for developing innovative models through observing variations across macroeconomic variables (Sadorsky, 2001). Explained the relationship between macroeconomic variables & oil companies stock price fluctuations by using the multifactor market model i.e. various market risk factors directly affect the price level of shares on index. CAPM also predicts the market returns through measuring risk factor hence it depicts that, required return of market increases simultaneously as the risk gets higher (Gay, 2016). Documented that increased level of oil prices is one of the major risk factor because it leads to inflationary pressure on oil consuming economy instead of oil producing economies (Mudhaf and Goodwin, 1993). Examine the two factor model & concluded that risk of oil price fluctuations with respect to NYSE returns are highly unpredictable. In the light of earlier mentioned literature the most confronted risk is the dynamics of macroeconomic variables.
Stock market is the fundamental factor for economy growth & provides a channel for FPI of different economy sectors. Additionally, FPI indicates the high investor confidence of our economy which is a positive sign for economic development. The responsiveness of market return fluctuation due to macroeconomic variables depend on the oil intensive or less oil intensive economies; (Waheed et al., 2018) heterogeneous market changes, the relationship & reaction of market becomes changed (Rahman and Mustafa, 2018). Furthermore, as we know raw material of industrial product is directly related to the world oil prices (Ayub, 2018) enlightens that high oil prices leads to high inflation in the economy, but it should be consider that oil prices fluctuations depends upon economy dynamics and varies from market to market. Whereas, another relevant risk dimension is exchange rate risk, which is crucially relevant in the context of Oil prices since the prices has been set internationally & the globally dominated currency is U.S. Dollars (Gomez and Zapatero, 2003).
Signaling theory, explains investment dynamics of two parties (personals or corporations) in the stock with respect to access and processing of different information by both parties. All market stakeholders requires information for making portfolio risk management, capital allocation & taking investment decisions, hence the information revealed in the market have a significant effect on their decision making process (Stiglitz, 2000). Highlighted that information asymmetries emerges when market participants don’t have a same sort of information & asymmetry occur when one participant hold a private information & doesn’t share to earn high level of profit. Signaling theory endorsed strong relationship with this study & illustrate that variations of oil prices, interest rate & exchange rate effects the stock market & also conveys a positive signs for the expectation of rise in inﬂation.
Efficient market hypothesis (EMH) also supports this study by concluding that asset prices set precisely in the market only when information flow is equally maintained between all market participants. In other words, material nonpublic information flow is entirely restricted. When market players forecast that there’ll be a rise in oil prices in next few months so, on this anticipation they’ll start adjusting their product price or cost and due to those adjustments there’ll not be a massive impact on firm market value. Oppositely, in case of high uncertainty regarding oil prices globally it’d be challenging to adjust their cost accordingly.
To investigate that relationship, regression model has been used for estimating the results.
For prognosis, following basic econometric model is specified in Equation 1:
LSRt = α1 + β1LOILt + β2LEXt  β3LINTt + β3LFDIt + et (1)
GARCH (1,1)
Mean Equation 2:
GARCH = C1 + C2*(IV) + e (2)
Variance Equation 3:
GARCH = C(2) + C(3)*RESID(1)^2 + C(4)*GARCH(1) (3)
Where, LSR = Log of Sector Return, LOIL = log of Oil, LEX = log of exchange rate, LINT = log of interest rate, LFDI = log of foreign direct investment, e = random error term & t = on specific time notation.
This study hypothesized positive relationship between oil prices & stock return of different sectors as other studies, Sadorsky (2001); Narayan and Narayan (2010); Faff and Brailsford (1999) documented that emerging markets stock markets were have a positive response against increases oil prices (Ma and Kao, 1990). And Mukherjee and Naka (1995) suggested that currency devaluation is favorable for export oriented economies but in Pakistan, export level is relatively low in comparison of import & on the other hand when dollar appreciates FPI losses their confidence & in result stock market effect significantly. The relationship between interest rate & stock market returns are hypothesized negatively because when required return of market increases, intrinsic value of stock reduces. Markowitz portfolio theory also suggested raise in risk – free rate of market leads to higher required return & due to that risk of the market increases. Moreover higher rates also provide the diversification opportunity for the investor.
Monthly Data of 13 years from 2005 to 2018 have been taken. Data for variables have been taken from World Development Indicators, investing.com, SBP, IMF & PSX websites. The data of number of tourist arrivals have been taken from Pakistan yearly statistical book.
Nonstationarity is the fundamental problem in the time series data & for resolving that problem multiple tools have been used in the empirical part of the study. Both dickey fuller & break point unit root tests endorse the nonstationarity in the data series of all variables at loglevel 1 & on 1st difference data series of variables become stationary. Long term relation between variables is checked through cointegration model & volatility of stock market with respect to macroeconomic variables is tested by using (GARCH (1,1)) model.
Results shows that at level, data series are nonstationary. The results of LSM reveal that stock market is 90.03% dependent on model’s variables & after adjusting into RSquare degree of influence of macroeconomic variables reduces to 89.78% which is exhibited in Table 1. LSM results shows that overall model has a significant impact on stock market return as Pvalue is less than 0.05 but has a weak relation with oil prices as the pvalue of oil prices is greater than 0.05, indicating that oil prices reduces model’s significance. Moreover, as per equation estimation statistics, stock market has a positive relationship with FDI & ER but negative with oil prices & interest rates. If one unit changes in independent variable, so in result, dependent variable changes β times with respective directions.
Table1. Least Square Method Statistics
Variable  Coefficient 
Std. Error 
tStatistic 
Prob. 
C  5776.580 
2710.825 
2.130931 
0.0347 
ERP  485.2710 
21.76666 
22.29424 
0.0000 
OPP  20.40958 
18.78535 
1.086463 
0.2789 
IRP  3010.880 
176.6644 
17.04294 
0.0000 
FDIP  1.184040 
0.300045 
3.946215 
0.0001 
Rsquared  0.900341 
FStatistics 
354.59169 

Adjusted Rsquared  0.897802 
Prob(FStatistics) 
0.0000000 

DurbinWatson stat  0.276283 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
Note: Least Square Model Test on EViews 9.
ERP stands for exchange rate price, OPP stand for oil prices, IRP stand for interest rate price and FDIP stand for foreign direct investment.
*Implies significance at 5% level.
In regression model, multicollinearity problem is often faced due to high correlation among the variables. Because of this problem, results could be regressed & anomalies may be encountered in the data analysis. To check multicollinearity problem variance inflation factor (VIF) model has been used & standard of VIF test is centered VIF values should be less than ten and this specifies existence of no association between model variables. See Table 2.
In time series data, correlation between data series observations also exist which is examined by employing serial correlation (LM) test & results reveal that autocorrelation problem exists in the model. In order to eliminate autocorrelation generalized LSM has applied. After application of this technique, outcome of serial correlation (LM) test Prob. value changed from significant to insignificant which means that association between variables has been removed & now the trend hasn’t regressed the results. See Table 2.
Table2. Variance Inflation Factor.
Variable  Coefficient 
Uncentered 
Centered 
Variance 
VIF 
VIF 

C  7348571. 
68.97221 
NA 
ERP  473.7874 
36.68348 
1.188881 
OPP  352.8894 
19.80822 
1.611540 
IRP  31210.30 
28.27969 
1.751652 
FDIP  0.090027 
2.366022 
1.044722 
BreuschGodfrey Serial Correlation LM Test:  
Fstatistic  0.780566 
Prob. F(1,154) 
0.3783 
Obs*Rsquared  0.806888 
Prob. ChiSquare(1) 
0.3690 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
VIF stand for Variance Inflation Factor.
*Variance Inflation Factor Test on EViews 9.
*Implies significance at 5% level.
Statistical data of variables have increasing trends & mostly time series data are nonstationarity. Graphs of data series is shows Figure 1.
Figure1. Nonstationary data Graphs.
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
*Graphical visualization through EViews 9.
For resolving the nonstationary problem in corresponding data set, a unit root test namely DickeyFuller Test is employed, which first checks existence of unit root on level 1 and then on first difference. The results of Table 3 indicate that all data series are nonstationary at level but on 1st difference data series become stationary. See Table 4.
When data series has a unit root at level one, it indicates existence of spurious regression and observations of variables itself predict the trend. This lead to spurious LSM results. When we applied the least square test on stationary data, model’s significance changed from significant to insignificant. The results are exhibited in Table 5.
When we eliminated the spurious regression from data series by unit root method, the dependence & coefficient of variables also changed. The results show that macroeconomic variables only predict 5.4% of stock market returns and 94.6% of the variation in stock market returns is explained by other factors & oil prices relation changed from negative to positive but still significant & coefficient of exchange rate changed from positive to negative but relation between variables is insignificant.
Table3. Level one DickeyFuller Unit Root Test.
Variables  tStatistic 
Prob.* 
Augmented DickeyFuller test statistic STMP  0.009110 
0.9573 
Augmented DickeyFuller test statistic ERP  0.154785 
0.9403 
Augmented DickeyFuller test statistic OPP  2.707716 
0.0749 
Augmented DickeyFuller test statistic IRP  1.11456 
0.7095 
Augmented DickeyFuller test statistic FDIP  4.974185 
0.0000 
*MacKinnon (1996) onesided pvalues.
Table4. 1st Difference DickeyFuller Unit root Test.
Variables  tStatistic 
Prob.* 
Augmented DickeyFuller test statistic (D)STMP  12.67820 
0.0000 
Augmented DickeyFuller test statistic (D)ERP  10.4053 
0.0000 
Augmented DickeyFuller test statistic D(OPP)  9.480470 
0.0000 
Augmented DickeyFuller test statistic D(IRP)  13.3565 
0.0000 
Augmented DickeyFuller test statistic D(FDIP)  14.04185 
0.0000 
*MacKinnon (1996) onesided pvalues.
Table5. Least Square Model.
Variable  Coefficient 
Std. Error 
tStatistic 
Prob. 
C  242.6474 
105.6272 
2.297206 
0.0229 
D(ERP)  87.99492 
80.27134 
1.096218 
0.2747 
D(FDIP)  0.002601 
0.124525 
0.020884 
0.9834 
D(IRP)  409.5354 
203.3633 
2.013812 
0.0457 
D(OPP)  34.70793 
15.24392 
2.276838 
0.0242 
Rsquared  0.054027 
Mean dependent var 
218.4069 

Adjusted Rsquared  0.029771 
S.D. dependent var 
1301.046 

S.E. of regression  1281.533 
Akaike info criterion 
17.18006 

Sum squared resid  2.56E+08 
Schwarz criterion 
17.27576 

Log likelihood  1377.995 
HannanQuinn criter. 
17.21892 

Fstatistic  2.227379 
DurbinWatson stat 
1.981972 

Prob(Fstatistic)  0.068506 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
The need of implementing unit root is to satisfy assumption of cointegration model. Cointegration test examines any possible long run relation among exchange rate, interest rate, oil prices & FDI from stock market returns.
Longrun relationship among the variables has been explored by employing johansen cointegration test. There is a possibility of long run association between stock returns & its anticipated determinants. Johansen cointegration technique has been proposed by Johansen and this methodology allows to empirically estimate strong longrun association. Johansen cointegration test has been based on Vector autoregressive model (VAR).
Johansen Cointegration method is constructed on maximum likelihood methodology and comprises of time trend which takes into account the probable outcome of a trending stock market returns. In cointegration model two test are used i.e. trace test & max statistic test. Trace test inspects the null hypothesis that the number of cointegrated trajectories in structure r, is less than or equal to r0 where r0<p & p is the number of determinants is the model, although alternative claim is that the impact matrix is a full rank. K max test inspects the null hypothesis that there are r0 integrated vectors against the alternative of r0 + 1 integrated vectors.
Results in Table 6 indicate the existence of cointegration in the relationship of stock market return, oil prices, exchange rate, FDI & interest rate.
Table6. Unrestricted Cointegration Rank Test (Trace).
Hypothesized 
Trace 
0.05 

No. of CE(s) 
Eigenvalue 
Statistic 
Critical Value 
Prob.** 
None * 
0.221502 
89.80219 
88.80380 
0.0423 
At most 1 
0.109566 
49.73990 
63.87610 
0.4254 
At most 2 
0.097621 
31.17252 
42.91525 
0.4342 
At most 3 
0.072048 
14.73724 
25.87211 
0.5964 
At most 4 
0.017183 
2.773166 
12.51798 
0.9023 
Trace test indicates 1 integrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon et al. (1999) pvalues 

Unrestricted Cointegration Rank Test (Maximum Eigenvalue) 

Hypothesized 
MaxEigen 
0.05 

No. of CE(s) 
Eigenvalue 
Statistic 
Critical Value 
Prob.** 
None * 
0.221502 
40.06229 
38.33101 
0.0313 
At most 1 
0.109566 
18.56738 
32.11832 
0.7610 
At most 2 
0.097621 
16.43528 
25.82321 
0.5062 
At most 3 
0.072048 
11.96408 
19.38704 
0.4185 
At most 4 
0.017183 
2.773166 
12.51798 
0.9023 
Maxeigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon et al. (1999) pvalues 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
In order to investigate volatility in the model, GARCH (1,1) model has been employed & in first step the breakpoint unit root test has been applied for checking the stationarity level in data series because first assumption of using GARCH model is data should be stationary on 1st difference so, as showed in Table 7, all data series become stationary at 1st difference & as per results there is no unit root in the data series p < 0.05.
Table7. Break Point Unit Root Test.
Variables  tstatistics 
Prob*. 
Augmented Dickey Fuller Statistic (DSTMP)  13.94 
< 0.01 
Augmented Dickey Fuller Statistic (DERP)  11.00 
< 0.01 
Augmented Dickey Fuller Statistic (DFDIP)  06.76 
< 0.01 
Augmented Dickey Fuller Statistic (DOPP)  03.82 
< 0.01 
Augmented Dickey Fuller Statistic (DIRP)  14.91 
< 0.01 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
The GARCH (1,1) is basically applied for capturing foremost features of time series data, such as stationarity by means of fat tails & volatility clustering. Furthermore, arch effect is completely opposite from the random walk concept. To analyze the model, all three GARCH (1,1) tests have been employed namely, correlogram q – statistics, correlogram squared residuals & heteroscedasticity. The correlogram qstatistics investigate the accuracy of equation which has been regressed on constant to check whether to modify or not, so as per result showed in Table 8, stock return, interest rate & FDI equations do not need to be modified on AR (1) process but data series of exchange rate & oil prices need to be modified through AR (1) process & after modification it’ll become insignificant.
To examine the presence of autocorrelation in the residuals Q – statistic test is employed. The above table shows the acceptance of the null hypothesis that states that there is no auto correlation in the time series data. The above correlogram of squared residuals test results direct that the residuals are not auto correlated as the p value is greater than 0.05 at all lags and now the series can be used for hypothesis testing and forecasting.
Table8. Residual Diagnostics tests.
Prob Values  
Correlogram QStatistics 
Correlogram Squared Residuals 

STMP 
ERP 
OPP 
IRP 
FDIP 
STMP 
ERP 
OPP 
IRP 
FDIP 
0.719 
0.285 
0.115 
0.515 
0.666 
0.861 
0.066 
0.362 

0.93 
0.268 
0.267 
0.468 
0.042 
0.807 
0.732 
0.473 
0.094 
0.657 
0.974 
0.004 
0.538 
0.2 
0.052 
0.934 
0.364 
0.599 
0.177 
0.629 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
*Implies test on Eviews 9 at 5% significance level.
Garch (1,1)
For estimating volatility across stock market & its respective elements (GARCH(1,1)) heteroscedasticity test is applied to get insight about the impact of macroeconomic variables on stock market returns. The (GARCH(1,1)) model has been applied for testing significant characteristics of time series data, such as stationarity by using fat tails and volatility clustering.
It is apparent from Table 9 mentioned below that macroeconomic variables have a positive coefficient from Karachi stock market except interest rate, that shows increasing trend in oil prices, exchange rate & FDI will cause high volatility in Karachi stock index but when interest rates increase, the behavior of index gets change & react negatively against interest rate fluctuations due to negative correlation. Furthermore, pvalue of all variables are less than 0.05 that shows that last trading day value doesn’t affect the next trading day. In addition to explain volatility, consider Resid(1) value that is also shows significance level that means volatility is effected from previous day residuals. Conditional variance is also effected from previous observation. Moreover, Best model for all the data series is (GARCH 1,1) because in this condition resid & Garch both coefficients are positive & sum of both coefficients is less than 1.
To inspect the existence of heteroscedasticity in the data series of the residuals, ARCH technique is applied & outcomes of test explain that there is no ARCH effect in the residuals. Out of five predictors three variables i.e. stock market, exchange rate & interest rate data series have an ARCH effect on level 1 but on 1st difference the effect has been removed through using AR(1) process but oil prices & interest rate data series don’t have an ARCH effect on level 1. In the light of final results of heteroscedasticity test we can say that there is no ARCH effect in the residuals. In other words, there is no heteroscedasticity in the data series; thus, the residuals can be said to be homoscedastic.
Table9. GARCH Results.
STMP  
Variable  Coefficient 
Std. Error 
zStatistic 
Prob. 
C  1.426245 
0.632079 
2.256435 
0.024 
Variance Equation  
C  2.932957 
2.028045 
1.446199 
0.1481 
RESID(1)^2  0.09001 
0.037169 
2.421676 
0.0154 
GARCH(1)  0.85086 
0.066028 
12.88643 
0 
ERP  
C  0.325047 
0.099278 
3.274099 
0.0011 
AR(1)  0.047224 
0.117129 
0.403184 
0.6868 
Variance Equation  
C  0.182709 
0.058713 
3.111896 
0.0019 
RESID(1)^2  0.380393 
0.145665 
2.611426 
0.009 
GARCH(1)  0.615416 
0.103787 
5.929576 
0 
FDIP  
C  0.325047 
0.099278 
3.274099 
0.0011 
AR(1)  0.047224 
0.117129 
0.403184 
0.6868 
Variance Equation  
C  0.182709 
0.058713 
3.111896 
0.0019 
RESID(1)^2  0.380393 
0.145665 
2.611426 
0.009 
GARCH(1)  0.615416 
0.103787 
5.929576 
0 
IRP  
C  0.04544 
0.381094 
0.11923 
0.9051 
Variance Equation  
C  5.803235 
2.835215 
2.046841 
0.0407 
RESID(1)^2  0.147826 
0.084069 
1.758385 
0.0787 
GARCH(1)  0.607372 
0.171051 
3.550819 
0.0004 
OPP  
C  2.230411 
1.218036 
1.831154 
0.0671 
AR(1)  0.02001 
0.182076 
0.10987 
0.9125 
Variance Equation  
C  25.08111 
40.67005 
0.616697 
0.5374 
RESID(1)^2  0.944249 
0.464674 
2.032066 
0.0421 
GARCH(1)  0.061886 
0.615844 
0.10049 
0.92 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
*Implies GARCH test on Eviews 9 at 5% significance level.
To inspect existence of heteroscedasticity in the distribution of the residuals, an ARCH test was implemented. Outcomes of ARCH test are portrayed in Table 10 for all the five variables. The ARCH test results specify that there is no ARCH effect in the residuals. In other words, there is no heteroscedasticity in data series; thus, the residuals can be said to be homoscedastic.
Table10. Results of Arch test.
Variables 
Fstatistic 
Obs*Rsquared 
Prob. F(1,158) 
Prob. ChiSquare(1) 
STMP 
0.41 
0.41 
0.52 
0.52 
ERP 
0.18 
0.18 
0.67 
0.67 
OPP 
0.03 
0.03 
0.86 
0.86 
IRP 
3.34 
3.31 
0.07 
0.07 
FDIP 
0.81 
0.81 
0.37 
0.37 
Source: World Development Indicators, Investing.com, State bank of Pakistan, International Monetary fund reports & PSX Website.
*Implies ARCH test on Eviews 9 at 5% significance level.
In order to analyze the moderating impact of FDI & FPI in the relation of macroeconomic variables & stock market returns, we used the regression model through estimating coefficient while making moderator through multiplying Mean equation where mean of data series become zero.
When FDI is introduced as a moderator in the model, then behavior of index returns against macro economic variables changes which is exhibited in Table 11 dependency of macroeconomic variables deviate significantly from 90% to 25% but significance of the model doesn’t change means with moderating impact model fitness is accurate, as PValues & coefficient of determinants has not changed except oil prices & exchange rate relation from index return i.e. without moderating impact, oil prices had negative relation which changed into positive due to moderating impact of FPI & exchange rate had a positive relation which then became negative.
Moreover, moderator in the model is unable to change the relation across the variables but overall dependency of index return changed significantly. See Table 12 & 13.
Table11. Model Summary.
Model 
R 
R Square 
Adjusted R Square 
Std. Error of the Estimate 
1 
.236a 
.056 
.032 
1280.31321 
a. Predictors: (Constant), MFDIP, OPP, IRP, ERP.
Table12. ANOVA.
Model 
Sum of Squares 
Df 
Mean Square 
F 
Sig. 

1 
Regression 
15119662.683 
4 
3779915.671 
2.306 
.051b 
Residual 
255715497.316 
156 
1639201.906 

Total 
270835159.998 
160 
Table13. Coefficients.
Model 
Unstandardized Coefficients 
Standardized Coefficients 
T 
Sig. 

B 
Std. Error 
Beta 

1 
(Constant) 
241.606 
105.506 
2.290 
.023 

ERP 
95.880 
81.436 
.093 
1.177 
.241 

OPP 
35.110 
15.201 
.182 
2.310 
.022 

IRP 
396.569 
203.362 
.154 
1.950 
.043 

MFDIP 
148.713 
272.542 
.043 
.546 
.586 
a. Dependent Variable: STMP.
When FPI is introduced as a moderator in the model, the behavior of index returns against macro economic variables changes which is shown in Table 14. Dependency of macroeconomic variables deviated significantly from 90% to 26% but significance of the model did not change, meaning with moderating impact model fitness is accurate, as PValues & coefficient of determinants has not changed except oil prices & exchange rate relation from index return i.e. without moderating impact oil prices had a negative relation which changed into positive due to moderating impact of FPI & exchange rate had a positive relation which became negative.
Moreover, moderator in the model is unable to change the relation across the variables but overall dependency of index return changed significantly. See Table 15 & 16.
Table14. Model Summary.
Model 
R 
R Square 
Adjusted R Square 
Std. Error of the Estimate 
1 
.267a 
.072 
.048 
1269.62897 
a. Predictors: (Constant), MFPIP, OPP, ERP, IRP.
Table15. ANOVA.
Model 
Sum of Squares 
df 
Mean Square 
F 
Sig. 

1 
Regression 
19369756.657 
4 
4842439.164 
3.004 
.020b 
Residual 
251465403.341 
156 
1611957.714 

Total 
270835159.998 
160 
a. Dependent Variable: STMP
b. Predictors: (Constant), MFPIP, OPP, ERP, IRP.
Table16. Coefficients
Model 
Unstandardized Coefficients 
Standardized Coefficients 
T 
Sig. 

B 
Std. Error 
Beta 

1 
(Constant) 
259.899 
105.086 
2.473 
.014 

ERP 
86.525 
79.496 
.084 
1.088 
.278 

OPP 
34.504 
15.059 
.179 
2.291 
.023 

IRP 
406.185 
200.384 
.158 
2.027 
.044 

MFPIP 
479.443 
279.647 
.132 
1.714 
.088 
a. Dependent Variable: STMP.
Panel estimation results shows in Table 17, full sample include Cement, Chemical Commercial banks, Engineering, power generation & distribution, fertilizer, food & personal care products, oil & gas marketing companies, oil &gas exploration companies, Pharmaceuticals & refinery. Panel cointegration method has been used for estimating long run relationship across macroeconomic variables & stock returns of different sectors. All data series are nonstationary at level 1 but become stationary on 1st difference which fulfill the basic requirement of panel data estimation. Three tests include pedroni, individual intercept & individual trend & nointercept & no trend test have been applied. The results show that nointegration exists in the model means no long term relation prevails between macroeconomic variables & stock returns of different sectors.
Table17. Kao Residual CoIntegration Test.
Measures  tStatistic 
Prob. 
ADF  0.332093 
0.3699 
Residual variance  3017.789 

HAC variance  1348.757 
Source: Panel data analysis on Eviews 9
Fixed & random effect model results show that exchange rate & oil prices have positive significant relation with sector wise stock returns but interest rate has negative impact on returns of sectors & dependency of returns is 17% & model is fit for analysis. See Table 18
Table18. Coefficient.
Variable  Coefficient 
Std. Error 
tStatistic 
Prob. 
C  168.9680 
63.57295 
2.657859 
0.0079 
ERP  2.210392 
0.239310 
9.236509 
0.0000 
IRP  21.94747 
1.980705 
11.08063 
0.0000 
OPP  0.416492 
0.210614 
1.977511 
0.0481 
Rsquared  0.167462 
Mean dependent var 
12.17287 

Adjusted Rsquared  0.166057 
S.D. dependent var 
169.1819 

S.E. of regression  154.4977 
Sum squared resid 
42440065 

Fstatistic  119.2126 
DurbinWatson stat 
0.132472 

Prob(Fstatistic)  0.000000 
Source: Panel data analysis on Eviews 9
The findings reveal that oil prices, exchange rate, interest rate & foreign direct investment are cointegrated with stock market return means long term relation has been exist across variables. As this result also validate the results (Narayan and Narayan, 2010) also found the strong long term association between the variables moreover, in order to validate the findings. LSM has been used for predicting the dependence level of index return on macroeconomic factors. Before removing spurious regression from model, dependence level of market return on its determinants were highly significant but when we eliminate spurious regression through using Log technique dependence level become change significantly. Exchange rate has statistically negative & insignificant relationship from index return which is also consistent from previous research findings because exchange rate fluctuation leads to high risk in the market & stock markets are conditionally risk sensitive (Abdalla and Murinde, 1997) also found the same results except in Philippines, due to change in investing behavior. FDI has very week relation with stock return which should be investigate further. In case of oil prices positive & significant relation has been exist which is impulsive with theoretical prospects (Phan et al., 2018). Documented, increase in oil price reduces market return & thus increase business cost, it effect negatively on stock prices but in Pakistan this relation behave differently due to low growth in industrial sector which leads to high goods demand. This is because when oil prices increases so overall economy price level increases which affect corporate profitability positively. Interest rate have negative relation with stock market & because risk free rate always provide a perfect hedge for stock market, it always been a negative correlation with index return but FDI has very week relation with stock return.
Panel regression model results explained that exchange rate & oil prices has positive significant impact on sector wise price change but interest rate has negative significant association which also similar to the findings of LSM test. Moreover, FDI & portfolio investment significantly moderates the relation of macroeconomic variables & index price variations. These results fulfill the extensive niche in the literature & via these findings investors & practitioners are now more equipped & able to take more concrete investing decisions while market confronted many internal or external challenges including vulnerability on foreign investment furthermore, also provide help to reveal insights in terms of investing behavior & also highlight the reasons of stock market volatility.
The goal is to first investigate the association between macroeconomic variables & stock market return volatility & then examine how moderators moderates the behavior of stock market with respect to its determinants. The sectoral basis analysis also run, to examine the association between macroeconomic variables & share price variations of 11 different PSX 100 index sectors. From last few years, declining trend had observed in PSX index 100 that is almost 26% to 30%. The contemporary variations of stock market has been captured by many economist, practitioners or analysts & urge to get it know what’s reasons of this drastic fluctuations which cost almost millions of dollar to the economy. After reviewing literature, got to know interest rate, exchange rate & specially oil prices are the main determinants of stock market but in case of Pakistan common perception & many analysts also believed that market is more dependent on sentiments & political influence instead of its determinants. So here’s the gap has been arises, moreover, also from last few years ratio of foreign investment increases significantly in different sectors of Pakistan economy due to CPEC & literature didn’t explained, how this phenomenon will transpose the relation across macroeconomic variables & stock market. For investigating the dependence of PSX return fluctuations on macroeconomic variables LSM has been used, cointegration model explained the longterm relation between variables & volatility in the data series were captured through (GARCH(1,1)) model. For estimating all the models, monthly time series & panel data were gathered for the period of 2005 to 2018.
Many factors still need to be addressed which can’t be explore in this study due to lack of technology or unavailability of data as Pakistan is a developing country & availability of historical data is one of the major concern. For further research, sectoral analysis could be conducted on other remaining sectors & also causality between the model variables needs to be analyze.
Funding: This study received no specific financial support. 
Competing Interests: The authors declare that they have no competing interests. 
Contributors/Acknowledgement: All authors contributed equally to the conception and design of the study 
Abdalla, I.S. and V. Murinde, 1997. Exchange rate and stock price interactions in emerging financial markets: Evidence on India, Korea, Pakistan and the Philippines. Applied Financial Economics, 7(1): 2535.Available at: https://doi.org/10.1080/096031097333826.
Aloui, C., D.K. Nguyen and H. Njeh, 2012. Assessing the impacts of oil price fluctuations on stock returns in emerging markets. Economic Modelling, 29(6): 26862695.Available at: https://doi.org/10.1016/j.econmod.2012.08.010.
Arouri, M.E.H., 2011. Does crude oil move stock markets in Europe? A sector investigation. Economic Modelling, 28(4): 17161725.Available at: https://doi.org/10.1016/j.econmod.2011.02.039.
Ayub, A., 2018. Volatility transmission from oil prices to agriculture commodity and stock market in Pakistan. Doctoral Dissertation, Capital University.
Broto, C., J. DíazCassou and A. Erce, 2011. Measuring and explaining the volatility of capital flows to emerging countries. Journal of Banking & Finance, 35(8): 19411953.Available at: https://doi.org/10.1016/j.jbankfin.2011.01.004.
Chen, N.F., T.E. Copeland and D. Mayers, 1987. A comparison of single and multifactor portfolio performance methodologies. Journal of Financial and Quantitative Analysis, 22(4): 401417.Available at: https://doi.org/10.2307/2330792.
Cong, R.G., Y.M. Wei, J.L. Jiao and Y. Fan, 2008. Relationship between oil price shocks & stock market: An empirical analysis from China. Energy Policy, 36(9): 35443553.Available at: https://doi.org/10.1016/j.enpol.2008.06.006.
Ewing, B.T., W. Kang and R.A. Ratti, 2018. The dynamic effects of oil supply shocks on the US stock market returns of upstream oil and gas companies. Energy Economics, 72: 505516.Available at: https://doi.org/10.1016/j.eneco.2018.05.001.
Ewing, B.T. and M.A. Thompson, 2007. Dynamic cyclical comovements of oil prices with industrial production, consumer prices, unemployment, and stock prices. Energy Policy, 35(11): 55355540.Available at: https://doi.org/10.1016/j.enpol.2007.05.018.
Faff, R.W. and T.J. Brailsford, 1999. Oil price risk and the Australian stock market. Journal of Energy Finance & Development, 4(1): 6987.
Ferrer, R., V.J. Bolós and R. Benitez, 2016. Interest rate changes and stock returns: A European multicountry study with wavelets. International Review of Economics & Finance, 44: 112.Available at: https://doi.org/10.1016/j.iref.2016.03.001.
Filis, G., S. Degiannakis and C. Floros, 2011. Dynamic correlation between stock market and oil prices: The case of oilimporting and oilexporting countries. International Review of Financial Analysis, 20(3): 152164.Available at: https://doi.org/10.1016/j.irfa.2011.02.014.
Gay, R.D., 2016. Effect of macroeconomic variables on stock market returns for four emerging economies: Brazil, Russia, India, and China. The International Business & Economics Research Journal, 15(3): 119126.
Gomez, J.P. and F. Zapatero, 2003. Asset pricing implications of benchmarking: A twofactor CAPM. The European Journal of Finance, 9(4): 343357.Available at: https://doi.org/10.1080/1351847021000025768.
Hamilton, J.D., 1983. Oil and the macroeconomy since world war II. Journal of Political Economy, 91(2): 228248.Available at: https://doi.org/10.1086/261140.
Javorci, B.S., 2004. Does foreign direct investment increase the productivity of domestic firms? In search of spillovers through backward linkages. American Economic Review, 94(3): 605627.Available at: https://doi.org/10.1257/0002828041464605.
Khan, K., S. ChiWei, A. Khurshid and A.U. Rehman, 2018. How often does the exchange rate granger cause the stock market in Pakistan? Applied Economics Journal, 25(1): 6578.
Khan, M.I., 2017. Falling oil prices: Causes, consequences and policy implications. Journal of Petroleum Science and Engineering, 149: 409427.Available at: https://doi.org/10.1016/j.petrol.2016.10.048.
Lee, Y.H. and J.S. Chiou, 2011. Oil sensitivity and its asymmetric impact on the stock market. Energy, 36(1): 168174.Available at: https://doi.org/10.1016/j.energy.2010.10.057.
Lipsey, R.E., R.C. Feenstra, C.H. Hahn and G.N. Hatsopoulos, 1999. The role of foreign direct investment in international capital flows. In International capital flows. University of Chicago Press. pp: 307362. Available from: http://www.nber.org/books/feld992 .
Ma, C.K. and G.W. Kao, 1990. On exchange rate changes and stock price reactions. Journal of Business Finance & Accounting, 17(3): 441449.Available at: https://doi.org/10.1111/j.14685957.1990.tb01196.x.
MacKinnon, J.G., 1996. Numerical distribution functions for unit root and cointegration tests. Journal of Applied Econometrics, 11(6): 601618.Available at: https://doi.org/10.1002/(sici)10991255(199611)11:6<601::aidjae417>3.0.co;2t.
MacKinnon, J.G., A.A. Haug and L. Michelis, 1999. Numerical distribution functions of likelihood ratio tests for cointegration. Journal of Applied Econometrics, 14(5): 563577.Available at: https://doi.org/10.1002/(sici)10991255(199909/10)14:5<563::aidjae530>3.3.co;2i.
Maghyereh, A., 2004. Oil price shocks and emerging stock markets: A generalized VAR approach. International Journal of Applied Econometrics and Quantitative Studies, 1(2): 2740.
MehrUnNisa, M.N. and M. Nishat, 2011. The determinants of stock prices in Pakistan. Asian Economic and Financial Review, 1(4): 276291.
Mudhaf, A.A. and T.H. Goodwin, 1993. Oil shocks and oil stocks: Evidence from the 1970s. Applied Economics, 25(2): 181190.Available at: https://doi.org/10.1080/00036849300000023.
Mukherjee, T.K. and A. Naka, 1995. Dynamic relations between macroeconomic variables and the Japanese stock market: An application of a vector error correction model. Journal of Financial Research, 18(2): 223237.Available at: https://doi.org/10.1111/j.14756803.1995.tb00563.x.
Narayan, P.K. and S. Narayan, 2010. Modelling the impact of oil prices on Vietnam’s stock prices. Applied Energy, 87(1): 356361.Available at: https://doi.org/10.1016/j.apenergy.2009.05.037.
Omay, T. and P. Iren, 2019. Behavior of foreign investors in the Malaysian stock market in times of crisis: A nonlinear approach. Journal of Asian Economics, 60: 85100.Available at: https://doi.org/10.1016/j.asieco.2018.11.002.
Oyerinde, A.A., 2019. Foreign portfolio Investment and srock market development in Nigeria. Journal of Developing Areas, 53(3): 110.Available at: https://doi.org/10.1353/jda.2019.0034.
Phan, D.H.B., V.T. Tran and D.T. Nguyen, 2018. Crude oil price uncertainty and corporate investment: New global evidence. Energy Economics, 77: 5465.Available at: https://doi.org/10.1016/j.eneco.2018.08.016.
Rahman, A.M. and M. Mustafa, 2018. Effects of crude oil and gold prices on US stock market: Evidence for USA from ARDL bounds testing. Finance and Market, 3(1): 19.Available at: https://doi.org/10.18686/fm.v3i1.1055.
Rehman, S., I.U. Chhapra, M. Kashif and R. Rehan, 2018. Are stock prices a random walk? An empirical evidence of Asian stock markets. Ethics, 17(2): 237252.Available at: https://doi.org/10.15408/etk.v17i2.7102.
Sadorsky, P., 2001. Risk factors in stock returns of Canadian oil and gas companies. Energy Economics, 23(1): 1728.Available at: https://doi.org/10.1016/s01409883(00)000724.
Sathyanarayana, S., S. Harish and S. Gargesha, 2017. Volatility in crude oil prices and its impact on Indian stock market evidence from BSE sensex. International Conference on Emerging Trends in Finance, Accounting and Banking, SDMIMD.
Stiglitz, J.E., 2000. Capital market liberalization, economic growth, and instability. World Development, 28(6): 10751086.Available at: https://doi.org/10.1016/s0305750x(00)000061.
Tahmoorespour, R., M. Rezvani, M. Safari and E. Randjbaran, 2018. The impact of oil price fluctuations on industry stock returns: Evidence from international markets. Journal of Management Info, 5(1): 112.Available at: https://doi.org/10.31580/jmi.v5i1.26.
Tsagkanos, A., C. Siriopoulos and K. Vartholomatou, 2018. Foreign direct investment and stock market development: Evidence from, a ‘new’ emerging market. Journal of Economic Studies.
Waheed, R., C. Wei, S. Sarwar and Y. Lv, 2018. Impact of oil prices on firm stock return: Industrywise analysis. . Emperical Economics, 55(2): 765780.Available at: https://doi.org/10.1007/s0018101712964.
Views and opinions expressed in this article are the views and opinions of the author(s), Asian Journal of Economic Modelling shall not be responsible or answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content. 