Dear STATA users,
I am very new to stata, so maybe this is a dumb question. I am trying to create a discrete choice model with cmmixlogit based on 444 routes who each have four alternatives. Any time i try to estimate more than four spatial variables in one model, i get the error that the input values are not feasible. It is even worse for interaction effects, sometimes i can't even estimate the interaction effect of one age variable on spatial variable on route choice. How do i solve this problem?
The code snippet that does work is cmmixlogit choice green_len aesthetic bridge_len waterperc light length, noconstant nolog or
However this does not: cmmixlogit choice aesthetic c.aesthetic#i.gender, noconstant nolog or
Related Posts with Question about numerical overflow
Standard Errors in the Frankel and Romer type study, using PPMLHi All, For those who are familiar with the Frankel and Romer (1999) study, the authors construct …
multiple rows loopDear stata members, I am trying to calculate the squared sum of all dyadic values in a group using …
How to expand results window?Good morning, I have item descriptions being cut off in results window, so getting an output with pa…
Nearest Neighbor Matching without Treatment Effects?I'm trying to do nearest neighbor matching in STATA 15 for Windows, but I'm just looking for matches…
Functional Forms of Independent VariablesHi, If our objective is to ascertain the relationship (specifically, sign and significance of Beta …
Subscribe to:
Post Comments (Atom)
0 Response to Question about numerical overflow
Post a Comment