To improve the efficiency of air quality analysis and the accuracy of predictions, this paper proposes a composite method based on Vector Autoregressive (VAR) and Random Forest (RF) models. In the theoretical section,...To improve the efficiency of air quality analysis and the accuracy of predictions, this paper proposes a composite method based on Vector Autoregressive (VAR) and Random Forest (RF) models. In the theoretical section, the model introduction and estimation algorithms are provided. In the empirical analysis section, global air quality data from 2022 to 2024 are used, and the proposed method is applied. Specifically, principal component analysis (PCA) is first conducted, and then VAR and Random Forest methods are used for prediction on the reduced-dimensional data. The results show that the RMSE of the hybrid model is 45.27, significantly lower than the 49.11 of the VAR model alone, verifying its superiority. The stability and predictive performance of the model are effectively enhanced.展开更多
金融稳定需要防范和化解金融市场之间的风险传染。与以往文献只是探究两个市场的风险传染不同,本文利用高维VAR for VaR模型将中国的汇市、债市、大宗商品、金融期货和股市等五个金融市场纳入统一框架,分析这5个金融市场在不同状态的风...金融稳定需要防范和化解金融市场之间的风险传染。与以往文献只是探究两个市场的风险传染不同,本文利用高维VAR for VaR模型将中国的汇市、债市、大宗商品、金融期货和股市等五个金融市场纳入统一框架,分析这5个金融市场在不同状态的风险溢出效应,这有助于捕捉冲击在不同金融市场之间传播而产生的间接影响。Wald检验和后验分析表明5个市场间只在危机或泡沫状态时存在明显的风险溢出效应。同时,本文利用压力测试发现单个市场的短期冲击影响会被其他金融市场如股市消化吸收,但4个金融市场都处于正常状态会明显降低其他金融市场如股市的左尾风险。此外,本文提出利用单个金融市场在同一时点的不同分位数计算每个金融市场在同一时点的预期收益、波动风险和崩盘风险,这种做法的好处在于结果更加稳健以及减轻极端值的影响。在此基础上,本文进一步探究金融市场间是否能够对冲彼此的波动风险和崩盘风险。结果显示大宗商品市场和金融期货市场能够有效地对冲其他金融市场的波动风险和崩盘风险,但汇市、债市和股市无法对冲其他金融市场的波动风险和崩盘风险。展开更多
文摘To improve the efficiency of air quality analysis and the accuracy of predictions, this paper proposes a composite method based on Vector Autoregressive (VAR) and Random Forest (RF) models. In the theoretical section, the model introduction and estimation algorithms are provided. In the empirical analysis section, global air quality data from 2022 to 2024 are used, and the proposed method is applied. Specifically, principal component analysis (PCA) is first conducted, and then VAR and Random Forest methods are used for prediction on the reduced-dimensional data. The results show that the RMSE of the hybrid model is 45.27, significantly lower than the 49.11 of the VAR model alone, verifying its superiority. The stability and predictive performance of the model are effectively enhanced.
文摘金融稳定需要防范和化解金融市场之间的风险传染。与以往文献只是探究两个市场的风险传染不同,本文利用高维VAR for VaR模型将中国的汇市、债市、大宗商品、金融期货和股市等五个金融市场纳入统一框架,分析这5个金融市场在不同状态的风险溢出效应,这有助于捕捉冲击在不同金融市场之间传播而产生的间接影响。Wald检验和后验分析表明5个市场间只在危机或泡沫状态时存在明显的风险溢出效应。同时,本文利用压力测试发现单个市场的短期冲击影响会被其他金融市场如股市消化吸收,但4个金融市场都处于正常状态会明显降低其他金融市场如股市的左尾风险。此外,本文提出利用单个金融市场在同一时点的不同分位数计算每个金融市场在同一时点的预期收益、波动风险和崩盘风险,这种做法的好处在于结果更加稳健以及减轻极端值的影响。在此基础上,本文进一步探究金融市场间是否能够对冲彼此的波动风险和崩盘风险。结果显示大宗商品市场和金融期货市场能够有效地对冲其他金融市场的波动风险和崩盘风险,但汇市、债市和股市无法对冲其他金融市场的波动风险和崩盘风险。
