Important: Read this before posting to this forum

  1. This forum is for questions related to the use of Apollo. We will answer some general choice modelling questions too, where appropriate, and time permitting. We cannot answer questions about how to estimate choice models with other software packages.
  2. There is a very detailed manual for Apollo available at http://www.ApolloChoiceModelling.com/manual.html. This contains detailed descriptions of the various Apollo functions, and numerous examples are available at http://www.ApolloChoiceModelling.com/examples.html. In addition, help files are available for all functions, using e.g. ?apollo_mnl
  3. Before asking a question on the forum, users are kindly requested to follow these steps:
    1. Check that the same issue has not already been addressed in the forum - there is a search tool.
    2. Ensure that the correct syntax has been used. For any function, detailed instructions are available directly in Apollo, e.g. by using ?apollo_mnl for apollo_mnl
    3. Check the frequently asked questions section on the Apollo website, which discusses some common issues/failures. Please see http://www.apollochoicemodelling.com/faq.html
    4. Make sure that R is using the latest official release of Apollo.
  4. If the above steps do not resolve the issue, then users should follow these steps when posting a question:
    1. provide full details on the issue, including the entire code and output, including any error messages
    2. posts will not immediately appear on the forum, but will be checked by a moderator first. This may take a day or two at busy times. There is no need to submit the post multiple times.

Could the LL of a full model worse than that for the reduced model?

Ask questions about the results reported after estimation. If the output includes errors, please include your model code if possible.
arohamirai
Posts: 18
Joined: 28 Jun 2020, 16:26

Re: Could the LL of a full model worse than that for the reduced model?

Post by arohamirai »

Dear Stephane,

Thank you very much for your thorough explanation. I believe this solves my problem, after accounting for the correlation between the h20 and h90 parameters, the LL improves compared with the basic model; this LL sits between the model using the same draw (perfect correlation) and that using two independent draws (no correlation), which makes sense theoretically. I feel I have learnt a lot here and thank you again.

Quick questions: in the correlated model you suggested, is preference dependence implied by the hinc_sig_20_90 parameter, e.g., a negative sign means that an individual who has a high sensitivity with 20% treatment would have a low sensitivity with 90% treatment? how do I calculate the correlation coefficient between the 20% and 90% parameters in Apollo? (especially calculating the standard error using Deltamethod?).

Thank you very much.

Tim
stephanehess
Site Admin
Posts: 974
Joined: 24 Apr 2020, 16:29

Re: Could the LL of a full model worse than that for the reduced model?

Post by stephanehess »

Tim

the sign of the correlation depends on the sign of the product of hinc_sig_20* hinc_sig_20_90. Section Section 9.2.5 in Kenneth Train's book. The Delta method can be used for this, but you'd need to work out the derivatives yourself - they're not implemented in Apollo for the standard errors.

Best wishes

Stephane
--------------------------------
Stephane Hess
www.stephanehess.me.uk
arohamirai
Posts: 18
Joined: 28 Jun 2020, 16:26

Re: Could the LL of a full model worse than that for the reduced model?

Post by arohamirai »

Hi Stephane,

Thanks a lot for your responses.

Tim
Post Reply