Hi
do you have some NA in your data? That would cause this problem
Stephane
Important: Read this before posting to this forum
- 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.
- 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
- Before asking a question on the forum, users are kindly requested to follow these steps:
- Check that the same issue has not already been addressed in the forum - there is a search tool.
- 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
- Check the frequently asked questions section on the Apollo website, which discusses some common issues/failures. Please see http://www.apollochoicemodelling.com/faq.html
- Make sure that R is using the latest official release of Apollo.
- Users can check which version they are running by entering packageVersion("apollo").
- Then check what is the latest full release (not development version) at http://www.ApolloChoiceModelling.com/code.html.
- To update to the latest official version, just enter install.packages("apollo"). To update to a development version, download the appropriate binary file from http://www.ApolloChoiceModelling.com/code.html, and install the package from file
- If the above steps do not resolve the issue, then users should follow these steps when posting a question:
- provide full details on the issue, including the entire code and output, including any error messages
- posts will not immediately appear on the forum, but will be checked by a moderator first. We check the forum at least twice a week. It may thus take a couple of days for your post to appear and before we reply. There is no need to submit the post multiple times.
Search found 1253 matches
- 13 Mar 2025, 09:45
- Forum: Errors during model validation and/or estimation.
- Topic: Error in LCCM using RP SP data
- Replies: 1
- Views: 1679
- 12 Mar 2025, 18:43
- Forum: Model estimation
- Topic: Latent Class model with EM algorithm
- Replies: 1
- Views: 1569
Re: Latent Class model with EM algorithm
Hi there
it looks like your EM estimation converged to a pretty bad solution. This is somewhat surprising, but EM does not in fact guarantee convergence to a better solution than classical estimation. Did you try some other starting values?
Stephane
it looks like your EM estimation converged to a pretty bad solution. This is somewhat surprising, but EM does not in fact guarantee convergence to a better solution than classical estimation. Did you try some other starting values?
Stephane
- 10 Mar 2025, 14:50
- Forum: Errors during model validation and/or estimation.
- Topic: Errors convergence latent class model
- Replies: 5
- Views: 9421
Re: Errors convergence latent class model
Hi
it looksl ike maybe you have a group of people here who always choose the SQ option. Could you look at the posterior class allocation probabilities and see whether the people very likely to fall into class b are also always choosing the SQ?
Stephane
it looksl ike maybe you have a group of people here who always choose the SQ option. Could you look at the posterior class allocation probabilities and see whether the people very likely to fall into class b are also always choosing the SQ?
Stephane
- 10 Mar 2025, 14:44
- Forum: Errors during model validation and/or estimation.
- Topic: MMNL model starting value cost variable
- Replies: 9
- Views: 14621
Re: MMNL model starting value cost variable
Asmita I don't have any ready made code for this specific case. What you would need to do is to take repeated draws of the model parameters (mu, sigma, etc) from the asymptotic parameter distribution, i.e. considering the covariance matrix, then product the draws for the random betas as you did, and...
- 10 Mar 2025, 14:43
- Forum: Model specification
- Topic: Parameter does not influence the log-likelihood of your model!
- Replies: 3
- Views: 8256
Re: Parameter does not influence the log-likelihood of your model!
Hi Gaby
only differences in utility matter. If you include e.g.
in both utilities, then it will simply drop out
Stephane
only differences in utility matter. If you include e.g.
Code: Select all
b_ht_stan*(ht_stan==2)Stephane
- 10 Mar 2025, 14:39
- Forum: Models and estimation
- Topic: Correlations across alternatives
- Replies: 7
- Views: 21954
Re: Correlations across alternatives
Hi what you really need to do here is a Cholesky transformation, as that would allow you to capture positive as well as negative correlation. Your specification is simply telling you there is correlation as the same sigma parameters are used in both utilities, so if multipled together they will be p...
- 07 Mar 2025, 20:29
- Forum: Post-estimation analysis/use of results
- Topic: delta method to calculate WTP (normal and log-normal parameters)
- Replies: 12
- Views: 33307
Re: delta method to calculate WTP (normal and log-normal parameters)
Hi
that's only easily possible if you can write down an analytic form for the WTP moments (as with a ratio of two lognormals). If not, you need something like in this paper DOI: 10.1016/j.trb.2013.09.010
Stephane
that's only easily possible if you can write down an analytic form for the WTP moments (as with a ratio of two lognormals). If not, you need something like in this paper DOI: 10.1016/j.trb.2013.09.010
Stephane
- 07 Mar 2025, 20:20
- Forum: Errors during model validation and/or estimation.
- Topic: MMNL model starting value cost variable
- Replies: 9
- Views: 14621
Re: MMNL model starting value cost variable
Asmita
that's only easily possible if you can write down an analytic form for the WTP moments. If not, you need something like in this paper DOI: 10.1016/j.trb.2013.09.010
Stephane
that's only easily possible if you can write down an analytic form for the WTP moments. If not, you need something like in this paper DOI: 10.1016/j.trb.2013.09.010
Stephane
- 27 Feb 2025, 17:37
- Forum: Model estimation
- Topic: Latent class model with covariates in class allocation model
- Replies: 2
- Views: 4884
Re: Latent class model with covariates in class allocation model
Hi
5 classes is almost always too many. Did you look at how the model fit changes when you increase classes?
Stephane
5 classes is almost always too many. Did you look at how the model fit changes when you increase classes?
Stephane
- 27 Feb 2025, 17:35
- Forum: Post-estimation analysis/use of results
- Topic: Correlation of MMNL random parameters calculated from unconditionals
- Replies: 3
- Views: 5673
Re: Correlation of MMNL random parameters calculated from unconditionals
Hi
no, you cannot reach that conclusion. You have not specified any correlation between the betas, so there can't be any. I assumed you had allowed for correlation, and wanted to see how strong it was. If you specify uncorrelated betas, they will be uncorrelated
Hope this helps
Stephane
no, you cannot reach that conclusion. You have not specified any correlation between the betas, so there can't be any. I assumed you had allowed for correlation, and wanted to see how strong it was. If you specify uncorrelated betas, they will be uncorrelated
Hope this helps
Stephane