基于生存分析的小汽车通勤者出发时刻研究

根据生存分析中的加速失效时间模型,对通勤者上班出发时刻选择行为进行研究分析.利用调查数据,建立了小汽车通勤者上班出发时刻选择的Weibull加速失效时间模型,得到了性别、年龄、家庭收入、单位上班时间和出行距离等是影响通勤者上班出发时刻选择的主要因素.结论表明模型的预测误差主要集中在10 min以内,模型的适用性和准确性较好,解决了Cox比例风险模型不能直接预测通勤者出发时刻的问题.

Car Commuters Departure Time Selection Analysis Based on Survival Analysis

The paper analyses the commuter departure time selection behavior according to the accelerated failure time model in survival analysis.This paper presents a Weibull accelerated failure time （AFT） model for car commuters departure time choices based on record data and receives that the main factors which influence the commuters departure time selection are sex,age,family income,attendance time and travel distance and so on.Specifically,some datasets are utilized to develop the prediction model and then,while others are employed to test the accuracy of prediction results of the model.The results show that the prediction errors of the model mainly range in time between 0 to 15 minutes and the model has a great applicability and accuracy.Compared with the Cox proportional hazards model,it solves the problem that the Cox hazards model can＇t predict directly the departure time.