Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Henrique Neder" <hdneder@ufu.br> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | RES: st: recovering markov transition matrix from cross sections surveys |
Date | Sun, 2 Jun 2013 10:56:59 -0300 |
Maarten Thanks for your suggestion but actually my information consists of a set of 10 independent random samples (not a panel) with information on the state of poverty (poverty or not) of each household and several independent variables that are supposed to explain (or cause) the poverty status. The sample at time t is not comprised of the same individuals that are contained in sample at time t-1. As there is not a panel, we cannot use the command xttrans to derive the transition probabilities. We cannot directly calculate the probability of an individual i in t-1 which is in the 0 state to pass, for example, to state 1 or remain in state 0. Therefore the suggestion of using a logit (or a set of logits) for estimating these transition probabilities based on independent samples and repeated in time is apparently not possible to be implemented. I imagine there might be a methodology that indirectly retrieve these transition probabilities using eg Bayesian statistics. Do you know any possibility for this or any literature that is driven in this way? Regards, Henrique -----Mensagem original----- De: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Em nome de Maarten Buis Enviada em: domingo, 2 de junho de 2013 05:35 Para: statalist@hsphsun2.harvard.edu Assunto: Re: st: recovering markov transition matrix from cross sections surveys You can get the raw transition matrix using -xttrans-. To make these transition probabilitities dpendent on covariates just estimate separate -(m)logit-s for each origin state. Things get more complicated when you want to model comon unobserved variables, that you will have to program yourself. Hope this helps, Maarten On Fri, May 31, 2013 at 7:31 PM, Henrique Neder <hdneder@ufu.br> wrote: > Hi, Stata listers > > Anyone know some econometric methodology (preferably with the > possibility of implementation in Stata) to retrieve (estimate) a > Markov transition matrix from microdata refering to a temporal > sequence (not too long) of cross sections surveys containing binary > (or nominal) information for some states.,eg poverty or not poverty > and other covariates (that possibly explain poverty or other category in any nominal variable)? > > Best > > Henrique Neder > > Professor associado - Instituto de Economia - Universidade Federal de > Uberlândia Av. João Naves de Ávila 2121, Bloco J, Campus Santa Mônica, > CEP: 38400-902 fone : 3230-9564 email: hdneder@ufu.br página: > www.ecn26.ie.ufu.br > > > > > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/statalist-faq/ > * http://www.ats.ucla.edu/stat/stata/ -- --------------------------------- Maarten L. Buis WZB Reichpietschufer 50 10785 Berlin Germany http://www.maartenbuis.nl --------------------------------- * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/