As of this writing, at least five programs and four estimation techniques* exist for estimating discrete-choice models. Until now, no tutorials existed on how to use this combination of tools, so we have attempted to fill this gap.
Common Data Set
A variety of software and estimation techniques populate this space, but many require intensive study to use. The animations that follow focus just on discrete-choice model estimation using a common data set that is also used to illustrate StatWizards' capabilities. For an introduction to the data set, watch our tutorial "Introduction to the Test Data Set". In addition to studying the following examples, we encourage you to download free versions of our products, go through the tutorials and make use of our innovative help systems.
To see an animated tutorial on estimating discrete-choice models, choose the program and estimation technique you want to explore from the links below.
|Building a CBC-HB model using Sawtooth Software||Program: CBC/HB||Technique: Hierarchical Bayes|
|Building a mixed logit model using Biogeme||Program: Biogeme||Technique: Mixed logit|
|Building a nested logit model using Biogeme||Program: Biogeme||Technique: Nested logit|
|Building a latent-class choice model using Latent GOLD||Program: Latent GOLD Choice||Technique: Latent class choice|
|Building a random parameters logit model using NLogit||Program: Limdep/NLOGIT||Technique: Mixed logit|
|Building a Random Parameters Logit model||Program: Limdep/NLOGIT||Technique: Nested logit|
|Building a mixed logit model using R's mlogit package||Program: R||Technique: Mixed logit|
|Building a nested logit model using R's mlogit package||Program: R||Techinque: Nested logit|