ApolloChoiceModelling forum

Ah, well that was silly. I don't think I had that issue with my earlier code. But it works now. Thanks for being the proofreader here.

I have pushed updates to the code and model output files at the GitHub link in the OP.

Thanks for your response. Interesting about the utility functions in lapply vs for. I just did a little test with the same model, and see that if I do V = list() ### compute class-specific utilities V[["walk_alone" ]] = ... Then the hessian is singular and no standard errors are returned. ...

I am estimating a multinomial logit model where observations have eight alternatives, and all are available. The data is available on GitHub https://github.com/urban-stack/kids-independent-travel/blob/main/data/all-purpose/usa-2017-all.rds , and the code to estimate the models is available on GitHub...

I wrote about this in another post that has yet to be moderated, but I can link to it there. There can be a problem identifying the length of unique ID's when the database object is an instance of class tibble. The error arises from this line in apollo_validateControls() length(unique(database[, apo...

Yes, I agree this looks very good! Getting tidy, glance, and augment methods written for apollo objects will help users access tools like modelsummary (which is amazing, and should be the model summary package to end all model summary packages), coefplot, and lots of other great pieces of the R ecos...

The apollo_saveOutput function does not allow the user to change which directory the model outputs get written to. This would be a very handy feature to allow users to integrate apollo into their existing data workflows. I would be happy to send a pull request with this improvement if I were pointed...

I'm in my first three hours of using apollo (outside of the examples). I'm trying to use it to estimate the following model from the mlogit package: library(mlogit) data("TravelMode", package = "AER") mlogit(choice ~ wait + vcost | size, TravelMode) I believe that I have finally ...

There are (at least) two different ways in which spatial correlation would show up in choice models. 1. Spatially correlated error between alternatives. This happens when there is some spatial attribute of the choice alternative, like in a destination choice model. That would seem to be handled with...

I have not done this, but am also very interested in knowing how it might be done.