These are replication files for Hamilton and Susmel(1994), "Autoregressive Conditional Heteroskedasticity and Changes in Regime,"
Journal of Econometrics, vol 64, pp 307-333. This is a slightly different setup from the SWARCH.RPF example file. In this case, the variance is g(S(t))h(t) where h(t) is a common ARCH process in the standardized squared residuals eps(t-k)^2/g(S(t-k)). The difference between this and the SWARCH example is that in the SWARCH (in effect) the divisor in the standardization is g(S(t)) rather than g(S(t-k)). In the Hamilton-Susmel model, this makes the likelihood at t dependent upon current and lagged regimes as in the MS-VAR. The SWARCH.RPF example actually uses the model as described in Cai(1994), “A Markov Model of Switching-Regime ARCH,”
Journal of Business and Economic Statistics, vol 12, no. 3, 309–316.
hs_garch.rpf- Standard ARCH/GARCH models
- (1.15 KiB) Downloaded 140 times
- hs_swarch_gl22.rpf
- Two regime-two lag model with Gaussian errors and asymmetry
- (4.2 KiB) Downloaded 156 times
- hs_swarch_tl32.rpf
- Three regime-two lag model with Student-t errors and asymmetry
- (4.38 KiB) Downloaded 143 times
- crspw.txt
- Data file
- (18.2 KiB) Downloaded 149 times
