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. 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.

Prediction with alternate dataset

Ask questions about post-estimation functions (e.g. prediction, conditionals, etc) or other processing of results.
Post Reply
mattwigway
Posts: 1
Joined: 08 Aug 2022, 17:36

Prediction with alternate dataset

Post by mattwigway »

Is it possible to use an estimated model object to predict for a new dataset (with the same schema)? In travel demand modeling we often will estimate a model using a survey dataset and then use the model to make predictions for a synthetic population. I don't see a way to specify a separate dataset in apollo_probabilities, and apollo_outOfSample seems to be only be used for cross-validation/train-test validation where a single dataset is split, not where there are separate datasets. Does Apollo have this functionality in another function I'm not aware of?
stephanehess
Site Admin
Posts: 1235
Joined: 24 Apr 2020, 16:29

Re: Prediction with alternate dataset

Post by stephanehess »

Hi

yes, this is straightforward.

Let's say that after model estimation, you want to use a new database called data_for_prediction.

you would then use these steps:

Code: Select all

database = data_for_prediction
apollo_inputs = apollo_validateInputs()
pred=apollo_prediction(model,apollo_probabilities,apollo_inputs)
The important step is the replacement of database by a new one, and then the call to apollo_validateInputs()

Stephane
--------------------------------
Stephane Hess
www.stephanehess.me.uk
work_996
Posts: 1
Joined: 14 Jan 2025, 11:06

Re: Prediction with alternate dataset

Post by work_996 »

Hello, I am encountering a similar issue and would appreciate your assistance. I estimated an MDCEV model using a dataset that includes independent and dependent variables. Now, I want to apply the estimated model to a new dataset, which only contains independent attributes, to predict the values of the dependent variables. After estimating the model, I ran the following code to make predictions:

new_data <- fread("new_data.csv")
database = new_data
apollo_inputs = apollo_validateInputs()
predictions_base = apollo_prediction(model, apollo_probabilities, apollo_inputs, prediction_settings=list(runs=30))

However, I encountered the following error:

Error in apollo_probabilities(apollo_beta, apollo_inputs, functionality = "prediction") :
object 'name_of_first_target_value' not found

It seems like the model is expecting some information related to the dependent values, but I am unsure how to resolve this. Any advice on how to properly prepare the new dataset or adjust the prediction process would be greatly appreciated! Thank you in advance!
stephanehess
Site Admin
Posts: 1235
Joined: 24 Apr 2020, 16:29

Re: Prediction with alternate dataset

Post by stephanehess »

Hi

you need the dependent variable to exist in the database as it is used in apollo_probabilities. Of course, if you make a prediction on a new dataset, what you can do is to just create a dummy version of that column. The values you put in it won't affect the predictions, you just need to ensure that the values you put in are in line with the model specification (e.g. in discrete choice, it has to be alternatives that are available, in MDCEV, the consumption has to sum to the budget)

Stephane
--------------------------------
Stephane Hess
www.stephanehess.me.uk
Post Reply