Abstract
This paper considers the design of a stated choice experiment intended to measure the marginal rate of substitution (MRS) between cost and an attribute such as time using a conventional logit model. Focusing the experimental design on some target MRS will bias estimates towards that value. The paper shows why this happens. The resulting estimated MRS can then be manipulated by adapting the target MRS in the experimental design.
Original language | English |
---|---|
Journal | Journal of Choice Modelling |
Volume | 16 |
Pages (from-to) | 43-49 |
Number of pages | 7 |
ISSN | 1755-5345 |
DOIs | |
Publication status | Published - 1 Sept 2015 |
Keywords
- C25
- C9
- Experimental design
- JEL C8
- Keywords:
- Misspecification
- Stated choice
- Willingness to pay