{"title":"COMEX 铜期货波动率预测:计量经济学模型和深度学习","authors":"Zian Wang, Xinyi Lu","doi":"arxiv-2409.08356","DOIUrl":null,"url":null,"abstract":"This paper investigates the forecasting performance of COMEX copper futures\nrealized volatility across various high-frequency intervals using both\neconometric volatility models and deep learning recurrent neural network\nmodels. The econometric models considered are GARCH and HAR, while the deep\nlearning models include RNN (Recurrent Neural Network), LSTM (Long Short-Term\nMemory), and GRU (Gated Recurrent Unit). In forecasting daily realized\nvolatility for COMEX copper futures with a rolling window approach, the\neconometric models, particularly HAR, outperform recurrent neural networks\noverall, with HAR achieving the lowest QLIKE loss function value. However, when\nthe data is replaced with hourly high-frequency realized volatility, the deep\nlearning models outperform the GARCH model, and HAR attains a comparable QLIKE\nloss function value. Despite the black-box nature of machine learning models,\nthe deep learning models demonstrate superior forecasting performance,\nsurpassing the fixed QLIKE value of HAR in the experiment. Moreover, as the\nforecast horizon extends for daily realized volatility, deep learning models\ngradually close the performance gap with the GARCH model in certain loss\nfunction metrics. Nonetheless, HAR remains the most effective model overall for\ndaily realized volatility forecasting in copper futures.","PeriodicalId":501084,"journal":{"name":"arXiv - QuantFin - Mathematical Finance","volume":"18 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"COMEX Copper Futures Volatility Forecasting: Econometric Models and Deep Learning\",\"authors\":\"Zian Wang, Xinyi Lu\",\"doi\":\"arxiv-2409.08356\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper investigates the forecasting performance of COMEX copper futures\\nrealized volatility across various high-frequency intervals using both\\neconometric volatility models and deep learning recurrent neural network\\nmodels. The econometric models considered are GARCH and HAR, while the deep\\nlearning models include RNN (Recurrent Neural Network), LSTM (Long Short-Term\\nMemory), and GRU (Gated Recurrent Unit). In forecasting daily realized\\nvolatility for COMEX copper futures with a rolling window approach, the\\neconometric models, particularly HAR, outperform recurrent neural networks\\noverall, with HAR achieving the lowest QLIKE loss function value. However, when\\nthe data is replaced with hourly high-frequency realized volatility, the deep\\nlearning models outperform the GARCH model, and HAR attains a comparable QLIKE\\nloss function value. Despite the black-box nature of machine learning models,\\nthe deep learning models demonstrate superior forecasting performance,\\nsurpassing the fixed QLIKE value of HAR in the experiment. Moreover, as the\\nforecast horizon extends for daily realized volatility, deep learning models\\ngradually close the performance gap with the GARCH model in certain loss\\nfunction metrics. Nonetheless, HAR remains the most effective model overall for\\ndaily realized volatility forecasting in copper futures.\",\"PeriodicalId\":501084,\"journal\":{\"name\":\"arXiv - QuantFin - Mathematical Finance\",\"volume\":\"18 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - QuantFin - Mathematical Finance\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.08356\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - QuantFin - Mathematical Finance","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.08356","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
本文利用计量经济学波动率模型和深度学习递归神经网络模型,研究了 COMEX 铜期货在不同高频区间的变现波动率预测性能。计量经济学模型包括 GARCH 和 HAR,深度学习模型包括 RNN(递归神经网络)、LSTM(长短期记忆)和 GRU(门控递归单元)。在采用滚动窗口法预测 COMEX 铜期货每日已实现波动率时,计量经济学模型,尤其是 HAR,总体上优于递归神经网络,其中 HAR 的 QLIKE 损失函数值最低。然而,当数据替换为每小时的高频已实现波动率时,深度学习模型的表现优于 GARCH 模型,而 HAR 达到了相当的 QLIKE 损失函数值。尽管机器学习模型具有黑箱性质,但在实验中,深度学习模型表现出了优越的预测性能,超过了 HAR 的固定 QLIKE 值。此外,随着每日已实现波动率预测期限的延长,深度学习模型在某些损失函数指标上逐渐缩小了与 GARCH 模型的性能差距。尽管如此,总体而言,HAR 仍然是预测铜期货每日已实现波动率最有效的模型。
COMEX Copper Futures Volatility Forecasting: Econometric Models and Deep Learning
This paper investigates the forecasting performance of COMEX copper futures
realized volatility across various high-frequency intervals using both
econometric volatility models and deep learning recurrent neural network
models. The econometric models considered are GARCH and HAR, while the deep
learning models include RNN (Recurrent Neural Network), LSTM (Long Short-Term
Memory), and GRU (Gated Recurrent Unit). In forecasting daily realized
volatility for COMEX copper futures with a rolling window approach, the
econometric models, particularly HAR, outperform recurrent neural networks
overall, with HAR achieving the lowest QLIKE loss function value. However, when
the data is replaced with hourly high-frequency realized volatility, the deep
learning models outperform the GARCH model, and HAR attains a comparable QLIKE
loss function value. Despite the black-box nature of machine learning models,
the deep learning models demonstrate superior forecasting performance,
surpassing the fixed QLIKE value of HAR in the experiment. Moreover, as the
forecast horizon extends for daily realized volatility, deep learning models
gradually close the performance gap with the GARCH model in certain loss
function metrics. Nonetheless, HAR remains the most effective model overall for
daily realized volatility forecasting in copper futures.