Loading...

A Simultaneous Approach for Activity Type and Activity Duration of Disabled People

Hajivosough, Shaghayegh Sadat | 2015

663 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47771 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Samimi, Amir
  7. Abstract:
  8. Daily activity patterns that include type and duration of activities are modeled simultaneously in MDCEV models. Conventionally, however, a diverse range of econometric models are used to sequentially model the decisions. Focusing on the disabled individuals, we aimed at comparing the prediction power of the two methods. A three level Nested Logit along with a Hazard model represent the sequential approach, the performance of which is compared to the MDCEV method. A random forest model is applied during the activity type prediction by MDCEV model. With considering 6.35 percent error in random forest model, the results confirmed that simultaneous approach improves activity type prediction by approximately 12 percent compared to the conventional approach. The results of prediction by hazard model demonstrate that 15.6 percent of people with disabilities need more than 24 hours to perform their daily activities, while this issue is addressed in the MDCEV model. Besides, influential variables are underscored and discussed, to pave the way for effective policies that encourage this vulnerable segment of population engage in various social activities in a way that reduces social exclusion and improve their quality of life
  9. Keywords:
  10. Daily Activity Pattern ; Hazard Index ; Disabled Individuals ; Three Level Nested Logit ; Multiple Discrete-Continuous Extreme Value Model (MDCEV)

 Digital Object List

 Bookmark

No TOC