本研究主要探討利用CARR模型與GARCH模型進行波動性預測及預測能力比較。並以天然氣價格、布蘭特原油現貨價格以及煤炭現貨價格對於二氧化碳現貨價格之波動性影響為主要研究對象,且以週變幅與週報酬率之形式進行各樣本之最適參數估計,以求最適樣本參數估計標準。再者,本研究也將以二氧化碳現貨價格作為主要研究對象,進行CARR模型與GARCH模型間樣本外預測能力比較,以驗證利用價格的動態極差結構之CARR模型是否較以價格報酬率為主之GARCH模型得到更精確之波動性預測。 實證結果顯示,在模型參數估計中,符合Chou (2005)之原則可達到最精確之參數估計;此外,利用CARR模型與GARCH模型均可證實,能源價格對於二氧化碳排放權價格皆具有影響性,且能源價格對於二氧化碳現貨價格皆存在槓桿效果。當上週能源相關新聞事件或負面消息因素產生時,將導致能源價格對於二氧化碳排放權現貨價格產生影響,且影響會持續至下一週,並間接造成二氧化碳排放權現貨價格之波動性增加。此外在預測能力之比較上而言,以CARR模型作為預測波動性之績效對於GARCH模型而言,相對較為精確。
This study employs the CARR (Conditional Auto Regressive Range) and GARCH models to predict the volatility of the EU Carbon Credit price and compares the prediction power of the two models. Based on various volatility estimation methods, this work examines the impact of natural gas price, Brent crude oil price and coal price on the volatility of EU Carbon Credit price. This investigation also attempts to obtain the assessing standard of optimal parameter estimation value by testing CARR and GARCH models using weekly range and weekly return. Additionally, this study compares the forecasting ability of CARR model with that of GARCH model on out-of-sample data by EU Carbon Credit price. The empirical results show that the estimated parameters obtained from CARRX(1,1) and GARCH(1,1) are the most concise ones, which confirms with the finding of Chou (2005). Furthermore, both the CARR model and GARCH model prove that Energy price has a significant impact on EU Carbon Credit price, and there is a leverage effect of energy price on EU carbon credit price. This study also finds that when negative energy news and other events occurred, the energy price influences the EU Carbon Credit price immediately, and the effect will lasts for one week. Moreover, the volatility of EU Carbon Credit price also increases indirectly. Regarding the forecasting ability on EU carbon credit price, this investigation finds that the accuracy and conciseness of the forecasting volatility implemented by CARR model outperforms that employed by GARCH model.