ApolloChoiceModelling forum

Hi

sorry, this paper was written long before Apollo so I don't have Apollo code.

Maybe you could have a go yourself and then show us your code and the errors you get, so we can help

Stephanea

hi

difficult to know without access to the code - can you share it?

but first, your model is clearly overspecified. You cannot estimate the parameters in both classes in the class allocation model. You fix delta_a, but you keep all the other parameters free

Stephane

Hi

looks correct to me. I would prefer specification 2 as you allow for further scale differences. In addition however, did you test for differences between the samples in how they react to the individual attributes?

Stephane

hi

I wonder if what is happening here is a version mix-up because some of your apollo function calls are using apollo:: and others are not.

Did you ensure you are running the latest version too?

Stephane

hi your attribute by attribute model basically failed immediately, i.e. did not estimate. I noticed you had an odd ] at the end of apollo_beta in one of your codes If you want me to debug this, then please send me your data and code and I'll have a look as it shouldn't afffect things at all Stephane

Hi simple fix. Have a look at your code. You have mnl_settings_RP$componentName = paste0("Class_",s) mnl_settings_SP$componentName = paste0("Class_",s) before you have: mnl_settings_RP = list( alternatives = c(local=1, metro=2, car=3, tw=4), ... so the list mnl_settings_RP does n...

David you are using randcoeff[["b_tc"]] = -exp( mu_log_b_tc + sigma_log_b_tc * draws_tc ) which means that the cost coefficient follows a negative lognormal distribution. A negative lognormal is a negative exponential of a Normal, so the parameters you are estimating are the mean and stand...

Jay

can you show us the output, please?

Thanks

Stephane

Hi I would in your case recommend option 1. Even when I treat income as a continuous variable in my models, I do not exclude people with missing data, nor do I assign them to another category. Rather, I estimate a different parameter for them, which would be in line with your idea to treat it as cat...

Hi I don't have any such example files, but like I said, this would all be accommodated by moving the apollo_panelProd call to be after apollo_lc. It's a very simply change for you to make For the class allocation model, you could then include covariates that vary across observations in the class al...