Talk:Mixed logit
This is the talk page for discussing improvements to the Mixed logit article. This is not a forum for general discussion of the article's subject. |
Article policies
|
Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL |
This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||||||||||||
|
Capitalization?
editIs Mixed Logit capitalized? It was used in every possible capitalization method in the article; in the interest of consistency, I capitalized all of them to "Mixed Logit". However, I cannot tell if this is correct. If it is, the page should be moved to Mixed Logit. Thank you. --lifebaka (Talk - Contribs) 18:16, 10 December 2007 (UTC)
This article may be too technical for most readers to understand.(September 2010) |
This is Wikipedia, not a tech manual. There really needs to be explaining, particularly in the intro to the article. Pharmboy (talk) 17:12, 11 December 2007 (UTC)
Correlation over time
editThe article makes a strawman comparison to standard logit in the "Correlation in unobserved factors over time" section. Although it's proper to explain how and why standard logit is problematic, it would be more useful to compare and contrast this technique with conditional logit (i.e., fixed-effects logit). Unfortunately, a search for conditional logit redirects to Discrete choice, where conditional logit is just a little technical blurb under Model F.
Related model - Logit Kernel
editThis article might benefit from a summary of some of the work done on the Logit Kernel model by Denis Bolduc, Moshe Ben-Akiva, Joan Walker and Michele Bierlaire. Logit Kernel starts from a different formulation for the utilities, but achieves the same flexible covariance structure.
Dr. Scarpa's comment on this article
editDr. Scarpa has reviewed this Wikipedia page, and provided us with the following comments to improve its quality:
My suggestion is two extend the page in two directions:
1) post-estimation, with the derivation of individual specific means and variances for random utility coefficients 2) WTP-space specifications, to avoid unbounded population distributions of marginal WTPs.
1) Reference to this can be found in Train (2003, 2009) chapter 11
2) Often analysts are interested in functions of random utility parameters, such as marginal WTPs, which are often ratios of random utility parameters. This creates problems as some of these ratios have distributions with poor empirical plausibility, such as undefined central moments. This can be circumvented by using utility specifications in the WTP-space, rather than in preference-space. The papers that address this issue are:
Train, K. & Weeks, M. (2005), Discrete Choice Models in Preference Space and Willing-to-pay Space, in R. Scarpa & A. Alberini, ed., 'Applications of simulation methods in environmental and resource economics', Springer Publisher, Dordrecht, The Netherlands, pp. 1-16.
Scarpa, R.; Thiene, M. & Train, K. (2008), 'Utility in WTP space: a tool to address confounding random scale effects in destination choice to the Alps', American Journal of Agricultural Economics 90, 994-1010.
Daly, A.; Hess, S. & Train, K. (2012), 'Assuring finite moments for willingness to pay in random coefficients models', Transportation 39(1), 19--31.
We hope Wikipedians on this talk page can take advantage of these comments and improve the quality of the article accordingly.
Dr. Scarpa has published scholarly research which seems to be relevant to this Wikipedia article:
- Reference : Ricardo Scarpa & Mara Thiene & Francesco Marangon, 2007. "Using flexible taste distributions to value collective reputation for environmentally-friendly production methods," Working Papers in Economics 07/24, University of Waikato, Department of Economics.