Loading...

Performance of two-stage trip generation models in estimating daily infrequent trips

Kermanshah, M ; Sharif University of Technology | 2003

127 Viewed
  1. Type of Document: Article
  2. Publisher: 2003
  3. Abstract:
  4. This study investigates validity of continuity assumption of error terms underlying multiple linear regression approach of trip generation models for infrequent trips. Since not many household members make daily trips for purposes like shopping or personal business, estimation of such trips with their discrete and non- negative nature and still close to zero averages, using conventional multiple linear regression, needs to be investigated. This study employs a two-stage trip generation model; the probability of making/not making trip is modeled first by binary probit, then, number of trips for those who have already made trips are estimated by the weighted least square method. The results are then compared with the results of single-stage regression model. The study findings show similarities in both sets of descriptive variables of the regression models. So, no justification was found for employing a very complex two-stage generation model
  5. Keywords:
  6. Binary Probit ; Infrequent trips ; Multiple linear regression ; Two-stage trip generation model
  7. Source: Amirkabir (Journal of Science and Technology) ; Volume 14, Issue 56 C , 2003 , Pages 1185-1198 ; 10150951 (ISSN)
  8. URL: https://www.sid.ir/paper/996/en