Skip to main content
Future Business Journal

Table 2 Serial correlation and conditional Heteroskedasticity tests results

From: Economic policy uncertainty and stock returns among OPEC members: evidence from feasible quasi-generalized least squares

Variable

Q = Stat

ARCH LM

k = 4

k = 8

k = 12

k = 4

k = 8

k = 12

EPU

247.39***

324.18***

359.79***

29.894***

9.659***

7.388***

Stock returns

 Algeria

0.21

0.908

0.925

0.008

0.009

0.01

 Ecuador

3.627

4.573

8.984

0.162

0.278

0.466

 Iran

35.797***

35.946***

36.106***

22.039***

12.343***

8.207***

 Iraq

0.056

0.104

0.148

0.0124

0.011

0.0097

 Kuwait

18.043***

20.052**

20.372*

3.371**

1.839*

1.129

 Nigeria

30.727***

35.071***

35.576***

8.217***

5.384***

3.541***

 S. Arabia

12.601**

23.672***

37.702***

3.466***

2.469**

2.289**

 UAE

33.865***

37.046***

52.483***

7.808***

5.345***

7.572***

 Venezuela

45.327***

60.727***

63.02***

17.919***

18.764***

12.187***

  1. The reported values for the serial correlation are the Ljung-Box Q-statistics, while for the heteroskedasticity test, this study uses the ARCH LM test F-statistics. Three (3) different lag lengths (k) of 4, 8, and 12 are considered for robustness. The null hypothesis for the autocorrelation test is that there is no serial correlation, while the null for the ARCH LM test is that there is no conditional heteroscedasticity. ***, ** and * denote significance at 1%, 5% and 10%
Back to article page