Dear Ox users,
i now use an Ox package: MSVAR, to implement markov-switching analysis.
I find the following results:
Nobs Prob. Duration
Regime 1 12.4 0.35 1.15
Regime 2 22.6 0.65 2.14
However, i do not know how to find:
1. the standard error of probability of regime 1 and regime 2, ie. the
standard error of 0.35 and 0.65
(logit p21 and logit p12 are relevant??) (Anyone can let me know what
are logit p21 and logit p12, please???)
2. And, i also find:
standard error (Reg.1) 0.0913
standard error (Reg.2) 0.2845
However, i cannot find the "standard errors" of 0.0913 and 0.2845 (the
two fig. are searched using numerical optimatization methods, hence i
think there should have variances for the two figures.)
ch(y,y)|s1 ch(y,y)|s2 are relevant???? What are they???
I feel sorry for disturbing all of you so much and i would be grateful
for your help very much.
thanks.
regards,
Kai-yin Woo
Assistant Prof.
Dept of Economics,
Hong Kong Shue Yan College,
Hong Kong.
|