Open stefancoe opened 6 years ago
@hanase- I was able to fix this issue in the workplace location model. For the network/logsum variables, I was using a workplace zone id variable created for the persons table, which is fine for the observed data but does not work for sampled data. I fixed this and re-ran estimation, which resulted in reasonable coefficients.
This is related to this issue.
Might need to post here to figure out.