ApolloChoiceModelling forum

Hi, Finding good starting values for eMDC si not easy. The model is quite sensitive to the values of delta (complementarity/substitution) parameters, so it cat easily fail for values that are not close to the optimum. The failure is caused because before starting estimation, Apollo adds a small valu...

Dear Chris, As a first step, I would recommend you not fixing the alpha values to zero. They usually tend towards zero, but fixing them at zero might cause problems. So I would recommend you fixing them to something very close to zero (from the right), but not quite zero, say 1e-9. Concerning the in...

Hi, I am afraid using the function "eval" inside apollo_probabilities, apollo_randCoeff, or apollo_lcPars will not work. Before running the model, Apollo does a series of modifications to these functions, to both optimise them and allow for the use of analytical gradients. That is why most...

Hi Asmita, First, I would like to say that I do not have much experience with effects coding. Personally, I always use dummy coding, so I will try to answer the best I can. Secondly, I want to highlight that identification of mixed mnl models with categorical variables is not simple, as the choice o...

Hi Asmita,

Could you also post the code for each model? Looking at least to the definition of the utilities might help us understand things better.

Best wishes,
David

Hi, I am not sure I fully understand your question. According to your equations, your latent variable is continuous, because you define it as: LV = b_female * Gender_female + b_higher_education * Education_higher_education + eta Do you mean that the indicators of your latent variable are dichotomous...

Hi, I had a quick look at the code, and I believe you should not comment out the line including n_classes inside apollo_inputs after calling apollo_validateInputs. Please modify the code as shown as follows. # Validate inputs apollo_inputs <- apollo_validateInputs() apollo_inputs$n_classes <- n_clas...

Hi Andrew, Out of the top of my mind, I don't have a reference for the issue with using the same starting values for latent classes (LC). The literature on identification of LC models is spread across multiple research areas and it is tricky business. However, imagine a model with two classes, only ...

Hi, We have come to realise that model in particular fails to converge on Windows, but it does converge on macOS. Beyond the OS differences, the issue is a very common one among MDCEV models with outside good using the alpha-gamma parametrisation, i.e. models where you have a common alpha among all ...

Hi, I modified your code by adding line 48 (inside apollo_lcPars). It is kind of a nasty trick, but it should allow you to run your model. I cannot test the code as I don't have the database, but let us know if it doesn't work. # Initialize Apollo apollo_initialise() # Set up core control parameters...