mixl-package: Estimate mixed multinomial logit models

In mixl: Simulated Maximum Likelihood Estimation of Mixed Logit Models for Large Datasets

Estimate mixed multinomial logit models

Description

Estimate mixed multinomial logit models using (simulated) maximum likelihood estimation. The package supports standard mnl, mixed-logit and hybrid choice. Using compilation to C++, model estimation is significantly faster than in native R code.

Details

This section should provide a more detailed overview of how to use the package, including the most important functions.

Author(s)

Joe Molloy <[e-mail]>.

References

Molloy, J., Becker, F., Schmid, B., & Axhausen, K. W. (2021). mixl: An open-source R package for estimating complex choice models on large datasets. Journal of choice modelling, 39, 100284.

Examples

  
  
  data("Train", package="mlogit")
  head(Train, 3)
  Train$ID <- Train$id
  Train$CHOICE <- as.numeric(Train$choice)

  mnl_test <- "
    ASC_B_RND 	= @ASC_B 	+ draw_2 * @SIGMA_B;

    U_A =             @B_price * $price_A / 1000 + @B_time * $time_A / 60 + @B_change * $change_A; 
    U_B = ASC_B_RND + @B_price * $price_B / 1000 + @B_timeB * $time_B / 60;
  "
  
  model_spec <- mixl::specify_model(mnl_test, Train)
  
  #only take starting values that are needed
  est <- stats::setNames(c(0,0,0,0,0,0), c("B_price", "B_time", "B_timeB", 
  "B_change", "ASC_B", "SIGMA_B"))
  
  availabilities <- mixl::generate_default_availabilities(Train, model_spec$num_utility_functions)
  
  model <- mixl::estimate(model_spec, est, Train, availabilities = availabilities, nDraws = 20)  
  
  summary(model)
  

mixl documentation built on May 29, 2024, 1:31 a.m.