Forecasting Stock Market Volatility: A Combination Approach
Table 2
In-sample estimation results.
In-sample estimation results for WTI
Coefficient
−0.793
0.470
0.166
0.077
−0.095
0.094
0.082
0.112
—
t-stat
−2.445
5.930
2.744
1.215
−1.261
1.384
1.530
2.308
—
In-sample estimation results for BRT
Coefficient
−0.846
0.487
0.166
0.077
−0.091
0.095
0.079
0.073
—
t-stat
−2.579
6.160
2.754
1.224
−1.189
1.374
1.472
1.659
—
In-sample estimation results for VIX
Coefficient
−9.820
0.151
0.044
0.069
−0.117
0.057
0.020
—
1.647
t-stat
−5.953
1.671
0.726
1.213
−1.673
0.864
0.386
—
5.719
In-sample estimation results for WTI + VIX
Coefficient
−9.475
0.131
0.047
0.061
−0.123
0.056
0.024
0.085
1.608
t-stat
−5.704
1.72
0.775
1.053
−1.676
0.874
0.455
1.820
5.593
In-sample estimation results for BRT + VIX
Coefficient
−9.663
0.134
0.045
0.059
−0.121
0.057
0.021
0.065
1.638
t-stat
−5.852
1.731
0.757
1.032
−1.668
0.861
0.396
1.771
5.690
Percent increase of
WTI
BRT
VIX
WTI + VIX
BRT + VIX
1.215
0.500
9.560
10.234
9.954
Note. This table reports the in-sample estimation results for the predictive regression models in (3), (4), and (5) for monthly stock volatility. We report the estimate of the slope coefficients, as well as the corresponding heteroskedasticity-adjusted t-statistic, based on the Newey–West method. We also show the percent increase in of the model of interest relative to that of the benchmark of AR in (2), The asterisks ,, and denote rejections of null hypothesis at 10%, 5%, and 1% significance levels, respectively.