LCLOGIT: Stata module to fit latent class conditional logit models via EM algorithm
Daniele Pacifico, Hong Il Yoo
Abstract
Daniele Pacifico, Hong Il Yoo
Abstract
lclogit fits latent class conditional logit models through an EM recursion proposed in Train (2008). The module can be useful to estimate nonparametric mixed logit models, as it allows to increase exponentially the number of mass points of each coefficient without compromising convergence.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
lclogit fits latent class conditional logit models through an EM recursion proposed in Train (2008). The module can be useful to estimate nonparametric mixed logit models, as it allows to increase exponentially the number of mass points of each coefficient without compromising convergence.
Key concepts: Mixed logit, Logit, Recursion (computer science), Nonparametric statistics, Latent class model, Logistic regression, Convergence (economics), Class (philosophy)