5-minute high-frequent data for SSE 50, CSI300, CSI500 and CSI 1000 indices

The realized recurrent conditional heteroscedasticity (RealRECH) model improves volatility prediction by integrating long short-term memory (LSTM), a recurrent neural network unit, into the realized generalized autoregressive conditional heteroskedasticity (RealGARCH) model. However, at present, there is no literature on the ability of the RealRECH model to fit and predict volatility in the Chinese market. In this paper, a study is conducted to test the in-sample explainability and out-of-sample prediction ability of the RealRECH model for the SSE50, CSI300, CSI500, and CSI1000 indices in the Chinese market and to determine whether it performs better than the RealGARCH model. The results of the in-sample analysis show that the RealRECH model not only provides better in-sample interpretability for all four indices but also captures the complex dynamics of time series volatility that the RealGARCH model cannot capture, such as long-term dependence and nonlinearity..., The high-frequency data in our research is from the Wind Financial Terminal at Southwestern University of Finance and Economics (SWUFE). SWUFE has purchased access to the Wind Database and has Wind terminals on campus, which allows us to download the data needed for our research from these terminals. Currently, there are no free public channels for accessing high-frequency data on Chinese stock indices. Researchers can either use the Wind Financial Terminal for paid access or purchase it through Chinese exchanges or brokers.
, , # 5-minute high-frequent data for SSE 50, CSI300, CSI500 and CSI 1000 indices

The objective of this study is to conduct in-sample analysis and out-of-sample prediction of the volatility of four Chinese stock indices using the RealGARCH model and the LSTM-RealGARCH(RealRECH) model, and to compare their effectiveness in analyzing and predicting the volatility of the Chinese stock indices. Therefore, we divided all data into in-sample and out-of-sample datasets.

The compressed package Data1 includes 2 folders, which are CSI300 and CSI500, and they both include 2 folders, SMC_for_RealGARCH and SMC_for_LSTM_RealGARCH.

The compressed package Data2&code includes 3 folders: CSI1000, SSE50 and RealRECH_norm, a compressed package RealRECH_norm and a file realized_china. The folders CSI1000 and SSE50 also include 2 folders, SMC_for_RealGARCH and SMC_for_LSTM_RealGARCH.The file realized_china contains the raw data we used in this study, which includes 5-minute high-frequency data for 2000 tradin...,

已实现递归条件异方差（RealRECH）模型通过将循环神经网络单元长短期记忆网络（LSTM）整合至已实现广义自回归条件异方差（RealGARCH）模型中，有效提升了波动率预测性能。但截至目前，尚无文献探讨RealRECH模型在中国市场的波动率拟合与预测能力。本文针对中国市场的上证50（SSE50）、沪深300（CSI300）、中证500（CSI500）及中证1000（CSI1000）指数，检验RealRECH模型的样本内解释能力与样本外预测能力，并验证其是否优于RealGARCH模型。样本内分析结果显示，RealRECH模型不仅对四只指数均具备更优异的样本内解释性，还能捕捉到RealGARCH模型无法捕获的时序波动率复杂动态特征，例如长期依赖性与非线性……本研究使用的高频数据取自西南财经大学（Southwestern University of Finance and Economics，SWUFE）的万得金融终端（Wind Financial Terminal）。西南财经大学已采购万得数据库的使用权限，并在校园内部署了万得终端，因此研究团队可通过该终端下载研究所需的全部数据。当前尚无免费公开渠道可获取中国股票指数的高频数据，研究人员需通过付费使用万得金融终端，或通过中国交易所、经纪商采购相关数据。
，，# 上证50、沪深300、中证500及中证1000指数的5分钟高频数据

本研究的核心目标为：采用RealGARCH模型与LSTM整合型RealGARCH（即RealRECH）模型，对四只中国股票指数的波动率开展样本内分析与样本外预测，并对比二者在分析及预测中国股指波动率方面的有效性。据此，我们将全部数据集划分为样本内数据集与样本外数据集。

压缩包Data1包含2个文件夹，分别为CSI300与CSI500，二者内部均设有2个子文件夹：SMC_for_RealGARCH与SMC_for_LSTM_RealGARCH。

压缩包Data2与代码文件包含3个文件夹：CSI1000、SSE50与RealRECH_norm，1个压缩包RealRECH_norm，以及1个名为realized_china的文件。其中CSI1000与SSE50文件夹同样包含SMC_for_RealGARCH与SMC_for_LSTM_RealGARCH两个子文件夹。文件realized_china包含本研究使用的原始数据，其中涵盖2000个tradin……的5分钟高频数据……