ApolloChoiceModelling forum

Thanks, David! Back when I asked this question, I wasn't fully aware of what I was asking about In any case, we figured it all out.

Thank you, Stephane, for your prompt response. In my model, I'm trying to apply direct categorization of individuals on the basis of latent variables approach as proposed by Bahamonde-Birke and Ortúzar here - https://bit.ly/3IeYJPQ . To answer your questions: 1. I want LVC1 to have a logistic error ...

Hi you would need to start with uniform draws in apollo_draws instead of Normal draws, and then use the appropriate formula to turn these into logistic draws inside apollo_randCoeff. Something like this: ### Set parameters for generating draws apollo_draws = list( interDrawsType="halton",...

Thank you for the reply. I reinstalled Apollo and stopped receiving the warning.

Hi forum admins, Apollo manual suggests setting workInLogs to TRUE for complex models (like Hybrid Choice). The way I read it that should work for data without the panel structure too. However, when I set it up, I get the message "Working in logs is only applicable with panel data; workInLogs s...

Thank you!

Hi, Thank you for bringing this to our attention. There is indeed a bug in Apollo v0.2.5 when estimating models with latent variables inside the class allocation. I have produced a new version without this problem that you can download from one of the links below, depending on your operation system...

Good morning Apollo forum moderators. Can you please help me to reflect in code the assumption that a random error term follows a logistic distribution and not a normal one? For example, how that can be coded in example #24 for "eta" (meaning that eta follows a logistic distribution). Than...

Hi Stephane, Thank you for your continuous enlightenment. Assuming that I still want to obtain allocation probabilities based on the estimates for latent variables without the choice component (for exploratory purposes). How can I code that in apollo? I tried adding to apollo_probabilities this: V=l...

Thank you, Stephane. Your comments and answers have finally guided me to what I'm after. To zoom out, I'm trying to classify the respondents in my dataset into distinct behavioural groups based on the estimates for their latent variables. To do that, I want to estimate the class allocation probabili...