A Method of Household Car Ownership Prediction Using Ordered Probit Model

Predicting the demand for cars accurately is of great significance to urban planning and parking facilities planning. This paper uses the data from the Beijing 2010 Resident Travel Survey and takes the family income level, family demographic characteristics, the geographical environment of the family and the personal attributes of the head of the household as parameters. Random forest method is used to rank the importance of explanatory variables and carry out variable reorganization. Use the ordered probability model (Ordered Probit) to establish the relationship between explanatory variables and the number of vehicles owned by the household. The model found that the three parameters of the number of family driver’s licenses, the level of family income, and the number of people have the greatest impact on the family’s decision to own a car, and the income group mean transformation has a better fitting effect.