ACCOMMODATING HETEROGENEITY AND HETEROSCEDASTICITY IN INTERCITY TRAVEL MODE CHOICE MODEL: FORMULATION AND APPLICATION TO HONAM, SOUTH KOREA, HIGH-SPEED RAIL DEMAND ANALYSIS

摘要: Multinomial logit models and nested are limited in that they cannot accommodate unobserved variations travelers' taste do not have flexible substitution patterns among alternatives because of the independence irrelevant property. Taking this background into account, traffic demand analysts recently used mixed model many studies. Unfortunately, most studies literature for joint analysis revealed-preference (RP) stated-preference (SP) data could simultaneously resolve two limitations just mentioned. The framework is to formulate an intercity travel mode choice RP-SP accommodates following behavioral considerations: (a) observed heterogeneity across individuals response level-of-service (LOS) factors, (b) heteroscedasticity alternatives, (c) scale differences between RP SP contexts. formulation estimated with maximum simulated likelihood method employs quasi-random Halton draws. applied examine behavior responses users HoNam, South Korea, high-speed rail changes conditions. empirical results show there a significant LOS attributes based on both individual characteristics. There improvement data-fit statistics when one introduces heteroscedasticity. These highlight need include within context modeling assist policy decision making.