Page 1 of 1

Markov switching model with time varying transition prob.

Posted: Sat Sep 26, 2015 8:57 am
by turana
Hello, I what to estimate the following two state model (carhart 4 factor model):


R= α + β ×〖MRK〗+ s ×〖SMB〗+ h ×〖HML〗+ p ×〖MOM〗+ e

where α, β, s, h, p, e are switching ( every coefficient and the variance).

I have successfully estimated the model with fixed transition probabilities, and the result are sensible. I also want to estimate this model with time-varying transition probabilities, where the transitions are explained by the following two equations:

P= ϕ( c d_1∆(CLI){2})
Q= ϕ(c d_2∆(CLI){2})

(CLI is the composite leading indicator - CLI is differenced and lagged two periods(months)

when I estimate the model, I get the transition probability (logit) coefficients of P11 (staying in regime 1 given that i am in regime 1) and P21 (transition from regime 2 into regime 1). The thing is that, I want the transition probability coefficient for P11(staying in regime 1 given that i am in regime 1) and P22 (staying in regime 2 given that i am in regime 2). IS there any whay to obtain these estimates, and how?

Appreciate all the help I can get 8)

Re: Markov switching model with time varying transition prob

Posted: Mon Sep 28, 2015 10:41 am
by EViews Glenn
P22 = 1 - P21