Abstract:
Considerable attention has been paid to the presence of non-response in large-scale travel surveys on the basis of which urban travel demand models are developed. In an earlier paper, the authors had shown that the effect of non-response can be reduced by careful model-building, with categorical trip-generation models as an example. In this paper, the authors extend the same philosophy to logit mode split models and exponential gravity models and show that the usual levels of non-response that one encounters in urban travel surveys have virtually no adverse effects on the parameter estimates of these models if the model has been correctly specified. Some simulation results are also presented to show the behavior of logit and exponential gravity model parameter estimates under conditions on non-response.
tr53.pdf