Forecasting Renminbi Exchange Rate Volatility Using CARR-MIDAS Model
Table 4
Diebold-Mariano statistics.
Horizon
CARR
GARCH-MIDAS
GARCH
1d
CARR-MIDAS
1.8760
8.3026
9.7510
CARR
7.6976
9.3290
GARCH-MIDAS
10.4591
2d
CARR-MIDAS
1.1952
6.1360
8.0666
CARR
5.7277
7.8760
GARCH-MIDAS
11.3301
3d
CARR-MIDAS
1.3302
5.7830
7.8312
CARR
5.3741
7.6804
GARCH-MIDAS
11.5681
4d
CARR-MIDAS
1.4811
5.4140
7.5467
CARR
4.9367
7.3325
GARCH-MIDAS
11.7271
1w
CARR-MIDAS
1.9078
5.0917
7.4106
CARR
4.4220
7.0158
GARCH-MIDAS
12.1817
2w
CARR-MIDAS
4.9525
3.8211
6.8328
CARR
1.9808
5.3129
GARCH-MIDAS
13.1066
1m
CARR-MIDAS
13.0779
4.3357
8.8228
CARR
-0.8880
4.4641
GARCH-MIDAS
15.1071
2m
CARR-MIDAS
15.0205
9.3679
14.9260
CARR
-1.4974
7.6723
GARCH-MIDAS
18.1445
3m
CARR-MIDAS
15.5806
8.9919
15.9034
CARR
-2.5555
9.3778
GARCH-MIDAS
18.7436
Note: A positive statistic indicates that the model in the row dominates the model in the column, and a negative statistic indicates that the model in the column dominates the model in the row. For the forecast horizons, d = day, = week, and m = month.
