Hi
I am trying to model migration within Sweden using a nested logit model and I have divided Sweden into 28 different regions. In the first step, I am modeling the choice to either stay in the region where individuals are currently residing or to move from the region as a function of indivudual characterists. In the second step, I am trying to model the choice of destination for those who moved as a function of characteristics at the destination areas.
When trying to run the model I get the "All constraints invalid message" and the program stops.
Because the stay-option is degenerate I use the constraint: constraint 1 [stay_tau]_cons=1
I have managed to make the nested logit model converge in other settings, so I am pretty sure that there is nothing wrong with how I enter the commands. I also employ the nlogitgen command before trying to run the model.
I am wondering whether this problem has anything to do with the fact that the sets of alternatives differs between indiviudals who move from different regions. For example, those who move from region number 2 can not chose to move to region 2 in the setup of destinations choices (because they are moving from region 2). And those who move from region number 3, can choose to move to region 2 (and to other regions) but not to region nr 3 (because thats where they are moving from.) Could this be the explanation?
Any input would be of great help!
Thanks!
Peter
Related Posts with problems with nested logit model
Histogram misbehavingI was plotting the simplest of histograms (observations by year) in Stata 14 and finding that the hi…
best package for matchingHi stata-users, I have been trying to understand psmatch2 but it hasn't been very helpful. I don't …
Multilevel model: how to determine whether there was a statistically significant change over timeHi all, I am running a simple multilevel model to determine whether a group of people had a statist…
Panel Data Estimation TechniqueHello Everyone, I am conducting some panel data analysis using fixed effects and system gmm. My tota…
Fixed effects - Unconditional quantile regressionHello, I´m running an unconditional quantile regression with country fixed effects. The command is:…
Subscribe to:
Post Comments (Atom)
0 Response to problems with nested logit model
Post a Comment