Investigation Spatial Transferability of Models Estimated on Stated Choice Data Collected on Four Different Continents

With the increasing costs of surveying, transferring model estimates obtained from one location and survey and applying them to another location is becoming increasing appealing. The transfer of previously estimated model outputs to new application contexts has the potential to reduce the need for new large scale data collection in the new application context as well as the reduction of the effort required to develop new models. Significant savings in cost and time can be achieved by applying this to other locations. However, the advantages in time and cost savings may be outweighed due to biases introduced if the transferred model does not adequately represent the behavior of individuals in the new application context. The majority of applications dealing with model transfer involve data collection within the same country or continent, or with relatively homogenous cultures. In this study, the authors compare the ability to transfer model outputs based on the same survey collected across four different continents, each with their own unique cultures and identities. The data used in this paper were collected in Australia, Chile, Taiwan and the United Kingdom. By using the same survey and survey processes across each location, the authors are able to discard effects related to the implementation of different survey instruments and/or interviewing process, and truly test whether model transfer can be achieved across such a diverse group of countries.The objective of this paper is to compare alternative methods of spatial transfer of discrete choice models across the four heterogeneous regions and cultures with particular interest being what outputs, if any, can be transferred. There are two main factors that impact upon the ability to transfer a model estimated from one set of data collected in one region and time, to another region and time. The first factor involves transfer bias, represented as differences between parameter estimates from the original model to corresponding coefficients associated with the new application context data. The second factor is that of statistical reliability. Although the second factor is of interest, it is with the first issue that the author concentrate in this paper. The authors are able to test what estimates may be transferred from one region to another.

• Record URL:
  • https://aetransport.org/past-etc-papers/conference-papers-pre-2009/conference-papers-2010
• Corporate Authors:

  Association for European Transport (AET)

  1 Vernon Mews, Vernon Street, West Kensington
  London W14 0RL,    
• Authors:
  • Chintakayala, P
• Conference:
  • European Transport Conference, 2010
  • Location: Glasgow Scotland, United Kingdom
  • Date: 2010-10-11 to 2010-10-13
• Publication Date: 2010

Language

• English

Media Info

• Media Type: Digital/other
• Pagination: v.p.
• Monograph Title: European Transport Conference, 2010 Proceedings

Subject/Index Terms

• TRT Terms: Alternatives analysis; Choice models; Data collection; Spatial analysis; Stated preferences; Statistical analysis; Surveying; Value of time
• Uncontrolled Terms: Discrete choice models
• Geographic Terms: Australia; Chile; Taiwan; United Kingdom
• Subject Areas: Planning and Forecasting; Transportation (General); I72: Traffic and Transport Planning;

Filing Info

• Accession Number: 01353922
• Record Type: Publication
• Files: TRIS
• Created Date: Oct 19 2011 12:52PM