Table 10: Estimating volatility dynamics in subsamples prior and post the change-points of
stock market returns indices.
|
SMI |
Process |
Subsamples |
k * Date |
Normal GARCH(1,1) Estimates | ||
|
ω |
α |
β | ||||
|
FTSE |
( rt )2 |
1-3338 |
0.003[4.409] |
0.067[10.46] |
0.919[111.8] | |
|
1-2484 |
13/7/98 |
0.003[3.669] |
0.058[6.744] |
0.915[64.18] | ||
|
2485-3338 |
0.017[2.839] |
0.100[4.410] |
0.845[22.40] | |||
|
HSI |
( rt )2 |
1-3338 |
0.016[15.80] |
0.124[19.29] |
0.855[128.6] | |
|
1-2248 |
15/8/97 |
0.023[18.71] |
0.134[17.04] |
0.810[97.42] | ||
|
2249-3338 |
0.029[4.387] |
0.092[7.462] |
0.878[63.48] | |||
|
NIKKEI |
( rt )2 |
1-3338 |
0.005[6.979] |
0.134[16.33] |
0.859[105.4] | |
|
1-966 |
16/9/92 |
0.002[2.206] |
0.229[11.11] |
0.801[58.14] | ||
|
967-2290 |
14/10/97 |
0.007[4.959] |
0.080[7.805] |
0.877[52.79] | ||
|
2291-3338 |
0.011[3.125] |
0.100[5.752] |
0.875[41.36] | |||
|
S&P500 |
( rt )2 |
1-3338 |
0.0007[4.712] |
0.039[12.979] |
0.958[294.4] | |
|
1-2291 |
15/10/97 |
0.0003[3.256] |
0.018[7.796] |
0.979[409.1] | ||
|
2292-3338 |
0.013[3.440] |
0.075[5.801] |
0.883[42.41] | |||
Note: The Moulines and Lavielle (2000) multiple breaks results in Table 7 for the absolute and squared returns processes are used to
create various subsamples of each stock market return index. The estimated Normal GARCH(1,1) coefficients as well as the
Power-ARCH coefficients are reported for the total sample (T=1-3338) as the various subsamples determined by the estimated break
points. Although not all subsamples have equal size some are approximately equal which allow for a better comparison of the estimated
parameters. The bold parameters emphasize the change in the size of the volatility estimates in most subsamples (especially the
parameters referring to the constant and ARCH effects of dynamic volatility).
34