Re: [Gretl-users] Markov-Switching Models