论文部分内容阅读
对金融资产收益波动率的研究是金融学理论的核心内容之一。本文研究了线性、非线性和长期记忆性等多种GARCH模型在不同分布设定下对碳期货波动率的预测效果,样本内的研究发现GARCH类模型能有效刻画碳期货收益率的条件异方差性,且t分布条件下拟合更好;外部冲击对碳期货收益波动率的影响具有持续性和非对称性。对样本外的研究发现,正态分布下的NAGARCH模型在预测碳期货波动率时是最优的。稳健性检验证明了结论是可靠的。
The study of the volatility of financial assets return is one of the core contents of financial theory. In this paper, we study the predictions of volatility of carbon futures under different distribution settings for a variety of GARCH models, including linear, nonlinear and long-term memory. The study in the sample shows that the GARCH model can effectively characterize the conditional heteroscedasticity of the yield of carbon futures And the fitting is better under the condition of t distribution. The impact of external shock on the volatility of carbon futures is persistent and asymmetric. Studies outside the sample found that the NAGARCH model under normal distribution is optimal for predicting the volatility of carbon futures. Robustness tests prove the conclusion is reliable.