此研究主要的目標是要對大型車的外部性這個假設以時間序列分析進行驗證，採用台灣從民國90年至101年的總體月時間序列資料。以車禍事故次數及死傷人數做為因變數，石油消費量及大型車的比例做為自變數，用以建立季節性IMA模型進行假設驗證。
定義排氣量1800CC以上的汽車為大型車，並將大型車的比例分為三種─(1)所有汽車中大型車之比例、(2)小貨車及小客車中大型車之比例和(3)小客車中大型車之比例。
最後得到的實證結果顯示，大型車的比例增加會顯著造成車禍事故件數的增加，對於死傷人數的增加較不顯著，其中又以所有汽車中大型車之比例之結果最為顯著。

The objective of this paper is to verify the hypothesis that increasing large automobiles cause the accident externality. This paper is done by time series analysis with aggregate time series data from Jan. 2001 to Dec. 2012 in Taiwan. Dependent variables are numbers of traffic accidents and number of injured passengers and fatalities which are examined in different models separately. Independent variables include gasoline consumption that was examined in Garcia-Ferrer and Hoyo(1987) and three different ratio of large automobiles examined in different models.
A large automobile is defined to be an automobile with engine displacement more than 1800CC. The externality is measured by three different large automobiles ratios which are the ratio of large automobiles in all automobiles, the ratio of large automobiles in all passenger sedans and pickup trucks, the ratio of large automobiles in all passenger cars.
The empirical result shows that the large automobile ratios significantly increase numbers of traffic accidents but less significantly increases number of injured passengers and fatalities. Among the three ratios, the ratio of large automobiles in all automobiles is more significant than the others.

摘要 I
Abstract II
目 錄 III
第一章 緒論 - 1 -
第一節 背景與研究動機 - 1 -
第二節 文獻回顧 - 1 -
第三節 論文架構 - 2 -
第二章 樣本資料及研究方法 - 3 -
第一節 樣本資料敘述 - 3 -
第二節 時間序列分析模型 - 5 -
第三章 實證結果及分析 - 7 -
第一節 季節性IMA模型建立 - 7 -
第二節 Garcia-Ferrer和Hoyo(1987)假設之模型驗證 - 8 -
第三節 本研究假設之模型驗證 - 9 -
第四章 結論 - 13 -
第一節 研究結論 - 13 -
第二節 改進方法 - 13 -
參考文獻 - 14 -
英文文獻 - 14 -
網站資料 - 14 -
附 錄 - 15 -
附表 - 15 -
附圖 - 18 -

英文文獻
1.Bair, S. T., Huang, R. J. and Wang, K. C., “Can vehicle maintenance records predict automobile accident?” Journal of Risk and Insurance, Vol. 79, Issue 2, PP. 567-584, 2012.
2.Box, G. E. P., and Jenkins, G. M., Time Series Analysis: Forecasting and Control, San Francisco: Holden-Day, 1976.
3.Garcia-Ferrer, A., and Del Hoyo, J., “Analysis of the Car Accident Indexes in Spain: A Mutiple Time Series Approach,” Journal of Business & Economic Statistics, 1987, Vol. 5, No.1, PP. 27-38.
4.Huang, R. J., Tzeng, L. Y., and Wang, K. C., “Heterogeneity of the Accident Externality from Driving,” Journal of Risk and Insurance, 2013 forthcoming.
網站資料
1.交通部交通統計查詢網，“交通事故”、“車輛登記數”，http://stat.motc.gov.tw/mocdb/stmain.jsp?sys=100
2.AREMOS台灣經濟統計資料庫「網路版」，“國內能源消費 - 石油產品 (公秉油當量)”，http://net.aremos.org.tw/
3.運輸安全網站資料系統，“民國100年度交通安全統計報表”，http://talas-pub.iot.gov.tw/PDF/mocka-24hr(100).pdf