應用計數模式分析機車肇事行為

本研究利用交通部運輸研究所90年度調查「我國機車持有及使用特性之研究」之問卷資料，針對外出活動使用機車之旅運者，在旅運過程中因不同之潛在影響因素而導致之交通肇事事故進行探討。應用一般化線性迴歸方法之卜瓦松、負二項、Zero-inflated及Hurdle迴歸模式建構肇事次數模式；又由於問卷數據中出現過多的0次資料，利用包含兩部份校估結果之Zero-inflated及Hurdle迴歸模式，可克服資料中出現過多0次肇事對於模式造成的偏誤，相較於基礎的卜瓦松及負二項之基礎計數模式具有較佳之配適度。本研究結果顯示，由第一部分危險駕駛行為下影響肇事機率模式顯著變數包括：個人平均月所得、年齡界於18-25歲之旅運者及每週車輛使用頻率六天以上且車齡高於七年以上等變數；其次，由第二部份肇事次數模式中可知：年齡界於18-25歲、個人平均月所得旅運者及車輛累積行駛里程數等變數均對於肇事機率有顯著的影響性。

關鍵字

機車意外事故；危險駕駛行為；計數模式； Zero-inflated, Hurdle模式

並列摘要

This paper investigates the factors influencing accident likelihood of motorcyclists in Taiwan by using the data collected by MOTC in 2001. Different models, including Poisson, Negative Binomial, Zero-Inflated Poisson and Hurdle are applied to estimate the models with dependent variable which is count data. Zero-inflated and Hurdle models with two parts are two cases of more general dual regime data generating process to overcome the problem when the number of zeros in the data exceed what would typically be predicted compared to Poisson and NB models. The results indicate thatthe effects of dangerous driving behavior on the accident likelihood appear to be significant in three aspects,” personal income”, “age” and “driving frequency over six days per week and vehicle age over seven years ”. The factors that influence the accident likelihood of motorcyclists are “personal income”, “age”, and “cumulative kilometers”.