An empirical test of a joint model of residential mobility and housing choice
A nested multinomial logit (MNL) model is used to estimate the joint choice of moving and housing selection. The coefficients are derived from a sequential application of the standard MNL model using maximum likelihood. The choice of the dwelling type is followed by choice of neighborhood and finally by the choice of moving or staying. The models are estimated for the household categories and nine neighborhoods with data from the Rand Corporation Housing Allowance Supply Experiment. The dwelling-type choice model fits reasonably well and shows expected significant impact of space on dwelling choice. The neighborhood choice model has smaller coefficients and is less successful. The alternative specific constants play a much larger role in determining the choice probabilities for neighborhoods. The final model of mobility choice exhibits a good fit for all the household categories. The results provide overall support for the sequential modelling approach outlined in this paper.
Year of publication: |
1985
|
---|---|
Authors: | Clark, W A V ; Onaka, J L |
Published in: |
Environment and Planning A. - Pion Ltd, London, ISSN 1472-3409. - Vol. 17.1985, 7, p. 915-930
|
Publisher: |
Pion Ltd, London |
Saved in:
freely available
Saved in favorites
Similar items by person
-
A multiple-attribute housing disequilibrium model of residential mobility
Onaka, J L, (1983)
-
Housing market search behavior and expected utility theory: 2. The process of search
Clark, W A V, (1982)
-
Clark, W A V, (1985)
- More ...