Macroeconomic indicators and their impact on stock markets in ASIAN 3: A pooled mean group approach

Figures & data

Table 1. Brief description of variables

Variables	Variables description
Dependent variable
Stock market (SM1)	Nifty
Stock market (SM2)	Shanghai stock exchange
Stock market (SM3)	Nikkei stock exchange
Independent variables
Exchange rate (ER1)	INR vs. USD
Exchange rate (ER2)	CNY vs. USD
Exchange rate (ER3)	JPY vs. USD
Consumer price index (CPI1)	CPI, India
Consumer price index (CPI2)	CPI, China
Consumer price index (CPI3)	CPI, Japan

Note: Overview of variables used in the study.

Table 2. Descriptive statistics

	Overview of Indian markets					Overview of Chinese markets					Overview of Japanese markets
	Nifty	Exchange rate (USD/INR)	CPI	Obs		Shanghai	Exchange rate (USD/CNY)	CPI	Obs		Nikkei	Exchange rate (USD/YEN)	CPI	Obs
Mean	6,021.46	54.25	49.45	107	Mean	2,706.75	6.49	52.56	107	Mean	12,789.11	97.35	43.54	107
SD	1,585.01	8.57	28.56	107	SD	606.85	0.29	29.84	107	SD	3,695.15	13.83	26.36	107
Min	2,755.10	39.28	1.00	107	Min	1,728.79	6.05	1.00	107	Min	7,568.42	76.23	1	107
Max	8,901.85	68.60	99	107	Max	4,611.74	7.18	104	107	Max	20,585.24	124.14	91	107
Skewness	0.16	0.12	0.0039	107	Skewness	0.97	0.24	−0.004	107	Skewness	0.46	0.22	0.11	107
Kurtosis	2.32	1.62	1.81	107	Kurtosis	3.88	1.74	1.82	107	Kurtosis	1.89	2.06	1.8	107

Figure 1. Graphical presentation of data-set.

Source: Stata 13.0.

Table 3. Augmented Dickey–Fuller unit root test: Model 1 (without first difference)

No of Observations: 107			Lags (0)
Exogenous: Constant
Dickey–Fuller test for unit root
Variables	ADF test	Null hypothesis	Result
Indian market
Nifty	−0.703*	Failed to reject h0	Variable is not stationary
CPI	−6.630	Reject H0	Variable is stationary
Exchange rate	−0.920*	Failed to reject h0	Variable is not stationary
Chinese market
Shanghai stock market	−3.060*	Failed to reject h0	Variable is not stationary
CPI	−2.437*	Failed to reject h0	Variable is not stationary
Exchange rate	−2.004*	Failed to reject h0	Variable is not stationary
Japanese market
Nikkei	−0.575*	Failed to reject h0	Variable is not stationary
CPI	−3.158*	Failed to reject h0	Variable is not stationary
Exchange rate	−0.898*	Failed to reject h0	Variable is not stationary

Note: A null hypothesis is accepted if test statistic < Critical value.

* Indicate acceptance of the null hypothesis of variable is not stationary at 1% significant levels.

Table 4. Augmented Dickey–Fuller unit root test: Model 2 (first difference)

No of observations: 107			Lags (0)
Exogenous: Constant
Dickey–Fuller test for unit root
Variables	ADF Test	Null Hypothesis	Result
Indian market
Nifty	−10.273	Reject H0	Variable is stationary
Exchange rate	−9.763	Reject H0	Variable is stationary
Chinese market
Shanghai stock market	−8.950	Reject H0	Variable is stationary
CPI	−17.592	Reject H0	Variable is stationary
Exchange rate	−7.282	Reject H0	Variable is stationary
Japanese market
Nikkei	−8.888	Reject H0	Variable is stationary
CPI	−8.808	Reject H0	Variable is stationary
Exchange rate	−9.002	Reject H0	Variable is stationary

Note: The null hypothesis is rejected if Test statistic > Critical value.

Table 5. Johansen’s cointegration test

	Pairs	Null	Trace statistics	Max statistics
			(Critical value)	(Critical value)
Indian Market (Lag 1)
1	Nifty—Exchange rate (INR vs. USD)	r = 0	9.31* (15.41)	9.31 (14.07)
		r = 1	0.00 (3.76)	0.00 (3.76)
2	Nifty—CPI (India)	r = 0	38.24 (15.41)	37.78 (14.07)
		r = 1	0.45* (3.76)	0.45(3.76)
Chinese market (lag 2)
1	Shanghai—Exchange rate (CNY vs. USD)	r = 0	15.13* (15.41)	11.90 (14.07)
		r = 1	3.23 (3.76)	3.23(3.76)
2	Shanghai - CPI (China)	r = 0	32.88 (15.41)	22.11(14.07)
		r = 1	10.77(3.76)	10.77(3.76)
Japanese market (Lag 1)
1	Nikkei—Exchange rate (Yen vs. USD)	r = 0	10.11* (15.41)	9.35 (14.07)
		r = 1	0.75 (3.76)	0.75 (3.76)
2	Nikkei—CPI (Japan)	r = 0	10.50* (15.41)	10.17 (14.07)
		r = 1	0.33 (3.76)	0.33 (3.76)

Notes: Null hypothesis is accepted, if Test statistic < Critical value, Critical value considered at 5%.

Table 6. Granger causality test

	Pairs	df	F. test	p-value	Conclusion
Indian market (lag 1)
1	Nifty does not Granger cause Exchange rate	1	8.86	0.00*
	Exchange rate does not Granger cause NIFTY	1	3.67	0.05**	Bidirectional relation exists
2	Nifty does not Granger cause Inflation (India CPI)	1	0.89	0.34
	Inflation (CPI) does not Granger cause Nifty	1	0.35	0.55	Relationship does not exist
Chinese market (Lag 2)
1	Shanghai stock exchange does not Granger cause exchange rate	2	0.27	0.75
	Exchange rate does not Granger cause Shanghai stock exchange	2	1.10	0.33	Relationship does not exist
2	Shanghai stock exchange does not Granger cause Inflation (China CPI)	2	5.38	0.00*
	Inflation (CPI) does not Granger cause Shanghai stock exchange	2	3.23	0.04**	Bidirectional relation exists
Japanese market (Lag 1)
1	Nikkei does not Granger cause Exchange rate	1	1.26	0.26
	Exchange rate does not Granger cause Nikkei	1	7.55	0.00*	Unidirectional Relation
2	Nikkei does not Granger cause Inflation (Japanese CPI)	1	0.00	0.95
	Inflation (CPI) does not Granger cause Nikkei	1	0.44	0.50	Relationship does not exist

* indicate rejection of null hypothesis at 1% significant levels.

** indicate rejection of null hypothesis at 5% significant levels.

Table 7. Pooled mean group (PMG)

Pooled mean group of 3 countries
Variables	Coef	Std. Error
Long-run coefficient
Exchange rate	1.4360* (0.00)	0.2760
Inflation (CPI)	−0.0743(0.11)	0.0473
Error correction term	−0.2184* (0.00)	0.0681
Short-run coef
Exchange rate	1.4614 (0.42)	0.0681
Inflation (CPI)	0.0451(0.13)	0.0299
Constant	0.9855** (0.04)	0.4878
Log Likelihood	275.1331	–
Hausman test	2.08 (0.35)	–

Estimations are done using (xtpmg) routine in Stata. The pooled mean group, all controlling for country and time effects, while the panel (LR) shows long-run effects. The second panel reports short-run effects (SR) and the speed of adjustment (ec). Monthly data 2008–2016. The total number of observation for each variable 321 (107 from each country for individual variable).

* indicates significant at 1%.

** indicates significant at 5%.

Table 8. Pooled mean group (PMG)

Pooled mean group for India			Pooled mean group for China			Pooled mean group for Japan
Variables	Coef	Std. Error	Variables	Coef	Std. Error	Variables	Coef	Std. Error
Long run-coefficient			Long run-coefficient			Long run-coefficient
Exchange rate	1.4360*(0.00)	0.2760	Exchange rate	1.4360* (0.00)	0.2760	Exchange rate	1.4360* (0.00)	0.2760
Inflation (CPI)	−0.0743(0.11)	0.0473	Inflation (CPI)	−0.0743(0.11)	0.0473	Inflation (CPI)	−0.0743(0.11)	0.0473
Error correction term	−0.3517*(0.00)	0.6310	Error correction term	−0.1760*(0.00)	0.0534	Error correction term	−0.1274* (0.01)	0.4923
Short-run coef			Short-run coef			Short-run coef
Exchange rate	4.7068 * (0.00)	0.9106	Exchange rate	−1.6767*(0.00)	0.2137	Exchange rate	1.3541* (0.00)	0.1588
Inflation (CPI)	0.1019*(0.00)	0.0356	Inflation (CPI)	0.0001*(0.01)	0.0284	Inflation (CPI)	0.0332*(0.00)	0.0079
Constant	1.9544*(0.00)	0.3989	Constant	0.6004**(0.03)	0.2760-	Constant	0.4018** (0.03)	0.3989
Log Likelihood	275.1331	–	Log Likelihood	275.1331	–	Log Likelihood	275.1331	–
Hausman test	2.08 (0.35)	–	Hausman test	2.08(0.35)	–	Hausman test	2.08 (0.35)	–

*, ** indicates significant at 1% and 5%. Estimations are done using (xtpmg full, to obtain each individual country results) routine in Stata. Pooled mean group, all controlling for country and time effects, while the panel (LR) shows long-run effects. The second panel reports short-run effects (SR) and the speed of adjustment (ec). Monthly data 2008–2016. The total number of observation for each variable 107 for each country. Hausman test is indicating that PMG is consistent and efficient estimation than MG (the p-value of Hausman test is >5% and hence we report only PMG results in the study).