A new baseline model for estimating willingness to pay from discrete choice models

By Richard T. Carson, Mikoaj Czajkowski. Journal of Environmental Economics and Management, 2013

87 citations recorded by the source corpus.

Normal 0 21 false false false PL X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:Standardowy; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin-top:0cm; mso-para-margin-right:0cm; mso-para-margin-bottom:10.0pt; mso-para-margin-left:0cm; line-height:115%; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-fareast-language:EN-US;} We show a substantive problem exists with the usual ratio of coefficients approach to calculating willingness to pay (WTP) from discrete choice models. Correctly calculated confidence intervals for WTP using this approach are shown to always be bounded by minus and plus infinity. A variant of this problem has long been recognized for mixed logit, and we show that it occurs even in simple models like the conditional logit that are often used as a baseline reference specification. The problem occurs because the estimated standard error for the cost parameter implies some possibility that the true value of the cost parameter is arbitrarily close to zero. We propose a simple yet elegant way to overcome this problem by reparameterizing the parameter on the (negative) cost variable to enforce the theoretically correct (and empirically almost always found) positive coefficient using an exponential transformation of the original parameter. This reparameterization enforces the desired restriction that no-part of the confidence region for original cost parameter spans zero. With it the confidence interval for WTP is now finite and well behaved. Our approach also has implications for the estimates of WTP location parameters. When the cost parameter has little density near zero, mean and median WTP are quite close. With substantial density near zero, the two measures of location will diverge. Our proposed model is a straightforward way to implement using readily available commercial software. It has the same log-likelihood value as the usual baseline logit model and we recommend its use as the new standard baseline reference model.

Subfields: Economics and Econometrics, General Decision Sciences

View Paper: https://www.semanticscholar.org/paper/053a59363e3774ce80651c6d73dc9735d7de72fc

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

  1. Kenneth Train
  2. Moshe E. Ben-Akiva
  3. Daniel McFadden
  4. Stephane Hess
  5. Chandra R. Bhat
  6. William H. Greene
  7. David A. Hensher
  8. Joan L. Walker

In General Decision Sciences

  1. Kenneth Train
  2. Daniel McFadden
  3. Moshe E. Ben-Akiva
  4. Stephane Hess
  5. Caspar G. Chorus

Citations in the corpus