logitr R Package - John Helveston

logitr: Fast Estimation of Multinomial and Mixed Logit Models with Preference Space and Willingness-to-Pay Space Utility Parameterizations

Summary

Project PI Helveston (2023) has developed an R package called logitr, which is a handy tools for Discrete Choice Modeling. This literature has proved the feasibility of the logitr package for fast maximum likelihood estimation of MNL (Multinomial Logit) and MXL (Mixed Logit) models, and has provided step-to-step instructions of how to reproduce and expand to other DCM cases. The logitr R package can handle both preference space and WTP (willingness-to-pay) space utility models, and can apply multi-start optimization loops for non-convex log-likelihood models.

This literature is the core reference of methodology of our project.

Results

The reproduction of the examples can be accessed here.

Gaps

PI John Helveston is in the process of developing a new R package for DCM. I’ll see how different they are and will shift to the new one if necessary.

Back to top

References

Helveston, John Paul. 2023. “Logitr: Fast Estimation of Multinomial and Mixed Logit Models with Preference Space and Willingness-to-Pay Space Utility Parameterizations.” Journal of Statistical Software 105 (February): 1–37. https://doi.org/10.18637/jss.v105.i10.