Forecasting Volatility of the Stock Market in Materials Sector via Climate Risk Online Concern: An Empirical Study Based on Multi-Class Machine Learning Models

Read full text Summarise with VIRAT
Authors: he jin zuo, Xie Chi, Xiaoxin Li
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study uses a dataset to analyze the impact of online climate risk concerns on stock market volatility in China's materials industry. The data sample period is uniformly from January 2011 to December 2024, mainly including three categories: daily closing prices of the CSI 300 Materials Stock Index, Baidu Search Index data for climate risk-related keywords, and macroeconomic predictors for China. Stock index and macroeconomic data are sourced from the Wind database, and Baidu Index data is sourced from the official Baidu Index website.In the data processing, monthly realized volatility was constructed based on daily stock price data, and the natural logarithm was further used as the core research variable. Baidu Index data was averaged monthly, and macroeconomic forecasting factors maintained their original monthly frequency. All variables were standardized as needed before modeling. All data were uniformly aligned along the time dimension, ultimately forming a time-series tabular dataset indexed by month. Each row corresponds to a monthly observation, and each column corresponds to a different research variable. There are no missing values ​​in the dataset. The relevant data files are stored in Excel format, and Python-based autoregressive benchmark model code (Jupyter Notebook format) is provided, which can be directly read and reproduced using common statistical software or a Python environment.

Fields of Study & Subjects

Financial Risk and Volatility Modeling Stock Market Forecasting Methods Market Dynamics and Volatility

Publication details

Journal / source
ScienceDB
Publication date
9 January 2026
Language
ZH
Document type
Dataset