A Comparative Study of VaR Estimation for Structured Products
Table 7
Summary of accuracy performance of selected time series models through various criterions.
Time series model
Backtesting
Multivariate extension of Diebold and Mariano test
RMSE
MAE
W
L
H
W
L
H
W
L
H
W
L
H
Panel A: RUD notes
AR()-ARCH(5)
NB
B
NB
NB
B
B
B
NB
NB
NB
B
NB
AR()-GARCH(1,1)
NB
B
NB
B
NB
NB
B
NB
B
NB
NB
B
AR()-GARCH(1,4)
NB
NB
NB
B
NB
NB
B
NB
B
NB
NB
B
AR()-EGARCH(1,1)
NB
NB
NB
NB
B
NB
NB
B
NB
B
NB
NB
Panel B: SPO notes
AR()-ARCH(5)
NB
B
B
NB
B
NB
B
B
NB
NB
B
B
AR()-GARCH(1,1)
NB
NB
NB
B
NB
B
B
B
B
B
NB
B
AR()-GARCH(2,5)
B
NB
B
B
NB
B
B
B
B
B
NB
B
AR()-EGARCH(1,1)
NB
NB
NB
NB
NB
NB
NB
NB
NB
NB
NB
NB
Panel C: DJQ notes
AR()-ARCH(5)
NB
NB
NB
NB
NB
B
NB
NB
NB
NB
NB
NB
AR()-GARCH(2,1)
B
NB
B
B
B
NB
B
NB
B
NB
B
B
AR()-GARCH(2,5)
B
NB
NB
B
B
NB
B
NB
B
NB
B
B
AR()-EGARCH(2,1)
B
NB
B
NB
NB
B
B
B
B
B
NB
NB
AR()-EGARCH(2,2)
B
NB
B
NB
NB
B
B
B
B
B
NB
NB
Note that W, L, and H stand for the whole period, the low oil price period, and the high oil price period, respectively. NB and B denote “not better” and “better”, respectively. RMSE and MAE demonstrate root mean squared errors and mean absolute errors, respectively.