本研究以石化產業為主題,包含塑膠、紡織、化學、橡膠四個子產業,以台灣上市公司為研究對象,研究期間從2000年1月1日至2012年12月31日,以變異數法─共變異數法、歷史模擬法及蒙地卡羅模擬法估計股價日報酬之風險值,並進行績效檢定,挑選出最適模型後,對風險值進行追蹤資料迴歸分析,檢測可能影響風險值之因素。 本研究之主要發現如下: (1) 在二項檢定下以歷史模擬法預測績效較佳。 (2) 在條件概似比並無最適模型。 (3) 獲利能力、償債能力、資本結構、經營能力、原油價格及匯率對於風險值均 有顯著影響,其中以營收成長率、總資產周轉次數變動率及匯率變動對於風 險值解釋能力較佳。
With the development and popularity of risk management, Value at risk (VaR) has been an useful tool for practical purpose. Comparing to the traditional volatility measurement, VaR is a better way to measure and quantifies the downside risk. This research focuses on Taiwan’s petrochemical-related industry including plastic, textile, chemical and rubber industries. Three main approaches are used to evaluate VaR: Variance-Covariance approach, Historical Simulation approach, and Monte Carlo Simulation approach. We tries to compare the forecasting performance of different VaR estimating models and investigate the factors influencing VaR. The conclusions of this research are as follows: (1) Historical simulation performs better under unconditional likelihood test. (2) There is no suitable model under conditional likelihood test. (3) VaR is significantly influenced by profitability, solvency, capital structure, efficiency, oil price and foreign exchange.