ApolloChoiceModelling forum

Hi, In an MDCEV model the amount consumed of each alternative (activity) is your dependent variable, so you cannot calculate the impact of a change in it.I understand you want to understand how a change in consumption in one alternative influences the others, but it cannot be done directly. The only...

Hi, While Apollo can manage to get analytical derivatives for utilities defined via loops, this functionality is mostly for simple loops. When you use conditionals statements (ifs) inside the loop, it is very likely that Apollo won't be able to get the analytical derivative of that. So it's good tha...

Hi Zhao, I am not very familiar with effects coding (I myself always use dummy coding), so not sure what to recommend about it. But if you are including price as a continuous variable in your utility, then its coefficients almost certainly should be negative. About the rho squared, I am afraid there...

Hi, Let's consider five types of distributions and their recommended starting values. # ################################################################# # #### DEFINE MODEL PARAMETERS #### # ################################################################# # ### Vector of parameters, including any ...

Hi, Usually, I recommend first estimating a simple MNL model and use those estimated parameters as starting values for the mu (mean) of the the mixed logit, while starting their sigma (s.d.) at zero. This is assuming the random coefficients in your mixed model follow a normal distribution. If the ra...

Happy to hear applying constraints worked out, and the new value positive value is close in magnitude to the original negative one.

Best wishes
David

Hi, As of today, the current version is v0.3.0. Y can update it using the following line of code: install.packages("apollo") The manual needs to be updated to the latest version, but it will be soon. Anyhow, the instructions to install (or update) Apollo remain the same, regardless what th...

Hi, I would recommend using an exploded logit but adding error components to introduce correlation between alternatives, approximate a nesting structure. The code below is a modification of example "EL" in the examples page , where I introduced correlation between alternatives 1 and 2. You...

Hi, Yes, you can estimate a zero inflated negative binomial model with latent variables in Apollo. But there is no pre-coded negative binomial model likelihood, so you would have to write the likelihood yourself. You can look at section 5.7 for a discussion on how to add new models to Apollo. Altern...

Hi, There is nothing wrong with your model. This is a consequence of how Apollo works. In Apollo, you usually define random components as follows: randCoeff[["beta"]] = mu + sigma*draws From a mathematical perspective, if we consider "draws" as a random variable following a symme...