epoh 发表于 2011-10-18 15:44
zhangtao兄今天看了下Oxmetrics6.2,New features in PcGive 13发现PcGive 13已经可以做 Regime Switching M ...
epoh老师,您好!
为什么我估计的结果与您的不一样?
---- OxMetrics 6.01 started at 09:21:32 on 20-Oct-2011 ----
GNP82.xls loaded from E:\Ox Packages\Ox-MSVAR\GNP82.xls
Ox Professional version 6.00 (Windows/U) (C) J.A. Doornik, 1994-2009
---- Switching - PcGive 1.0 session started at 9:22:00 on 20-10-2011 ----
No Y variable selected
Failed to get data.
Switching( 1) Modelling DUSGNP by MS(2)
The dataset is: E:\Ox Packages\Ox-MSVAR\GNP82.xls
The estimation sample is: 1952(2) - 1984(4)
Coefficient Std.Error t-value t-prob
Constant(0) -0.494281 0.3262 -1.52 0.132
Constant(1) 1.09213 0.1306 8.36 0.000
sigma 0.822486 0.06252 13.2 0.000
p_{0|0} 0.676781 0.1289 5.25 0.000
p_{0|1} 0.0969262 0.04779 2.03 0.045
log-likelihood -184.929796
no. of observations 131 no. of parameters 5
AIC.T 379.859592 AIC 2.89969155
mean(DUSGNP) 0.719835 var(DUSGNP) 1.12849
Linearity LR-test Chi^2(3) = 17.738 [0.0005]** approximate upperbound: [0.0026]**
Transition probabilities p_{i|j} = P(Regime i at t+1 | Regime j at t)
Regime 0,t Regime 1,t
Regime 0,t+1 0.67678 0.096926
Regime 1,t+1 0.32322 0.90307
Used uniform probabilities to start recursion
Std.Error based on numerical Hessian matrix
SQPF using analytical derivatives (eps1=0.0001; eps2=0.005):
Strong convergence
Used starting values:
-0.11240 1.5649 0.65209 0.50000 0.50000