台灣酒駕被取締次數與總體經濟變數分析

Abstract

由於近年來酒駕被取締次數仍居高不下，故本研究利用台灣2003年1月至2020年12月內政部警政署之酒後駕車被取締件數月資料，探討各個數變數對酒駕被取締次數之影響及變數間相關性，找出可能影響酒駕之因素。根據單根檢定以及共整合檢定之檢定結果選用向量誤差修正模型進行迴歸分析，並以Granger因果檢定、衝擊反應函數以及變異數分解，來探討變數之間的關係。研究結果顯示，每人可支配所得中位數、酒類消費量及男性失業率對酒駕被取締次數之影響為正，酒類物價指數及女性失業率對酒駕被取締次數之影響為負。本研究針對以上結果提出相關政策方向建議，以供政府相關單位在制定酒駕防制相關政策時參考。

As the frequency of drunk driving has remained high in recent years, this study collected the number of drunk driving arrested from the National Police Agency's road traffic accident investigation data from January 2003 to December 2020. And used it to analyze the influence of various macroeconomic variables on the number of drunk driving arrested and the correlation between these variables, in order to find out the factors that may affect drunk driving. According to the results of Unit Root Test and the Cointegration Test, we use the Vector Error Correction Model for regression analysis, and use Granger causality test, impulse response function and variance decomposition to explain the relationship between variables. The final results shows that the median disposable income per person, alcohol consumption and male unemployment rate have a positive impact on the number of drunk driving arrested, while the alcohol price index and female unemployment rate have a negative impact on the number of drunk driving arrested. In view of the above results, this study puts forward relevant policy recommendations for reference by government units when formulating policies related to drunk driving prevention.