Table 7.
The result of alternative lags of OVX with the MCS test.
Forecasting models | QLIKE |
MSE |
|||
---|---|---|---|---|---|
Range | SeimQ | Range | SeimQ | ||
Panel A: h = 1 | |||||
MIDAS-RV | 0.0859 | 0.0528 | 1.0000 | 1.0000 | |
MIDAS-RV-OVX | 1.0000 | 1.0000 | 0.9922 | 0.9901 | |
MS-MIDAS-RV | 0.0859 | 0.0596 | 0.8269 | 0.9213 | |
MS-MIDAS-RV-OVX | 0.2020 | 0.2020 | 0.9922 | 0.9901 | |
Panel B: h = 5 | |||||
MIDAS-RV | 0.1107 | 0.0492 | 0.4408 | 0.2861 | |
MIDAS-RV-OVX | 0.1107 | 0.0492 | 0.4408 | 0.2427 | |
MS-MIDAS-RV | 0.1107 | 0.0682 | 0.4408 | 0.2960 | |
MS-MIDAS-RV-OVX | 1.0000 | 1.0000 | 1.0000 | 1.0000 | |
Panel C: h = 20 | |||||
MIDAS-RV | 0.0058 | 0.0144 | 0.0163 | 0.0544 | |
MIDAS-RV-OVX | 0.0069 | 0.0202 | 0.0163 | 0.0518 | |
MS-MIDAS-RV | 1.0000 | 1.0000 | 1.0000 | 1.0000 | |
MS-MIDAS-RV-OVX | 0.7455 | 0.7455 | 0.3898 | 0.3898 |
Notes: This table exhibits models’ MCS p-values with alternative lags of OVX, more specifically, we choose the two lags of OVX. The forecasts for different horizons of h = 1, 5, 20 are shown in Panel A, B, C. In this table, the MCS test p-value of 1indicates a better prediction.