Networking Your Way to a Better Prediction: Effectively Modeling Contingent Valuation Survey Data

The purpose of this paper is to empirically compare the out-of-sample predictive capabilities of artificial neural networks, logit and probit models using dichotmous choice contingent valuation survey data. The authors find that feed-forward backpropagation artificial neural networks perform relatively better than the binary logit and probit models with linear index functions. In addition, guidelines for modeling contingent valuation survey data and how to estimate median WTP using artificial neural networks are provided.


Issue Date:
2003
Publication Type:
Conference Paper/ Presentation
PURL Identifier:
http://purl.umn.edu/22152
Total Pages:
39
Series Statement:
Selected Paper




 Record created 2017-04-01, last modified 2017-12-06

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)