Can Renewable Energy Boost the Business Outcomes of Power Suppliers? An Empirical Study on Japan’s Power Producers and Suppliers

The data sources used to generate the cleaned CSV datasets are listed above. The final cleaned dataset used in the paper is included in CSV format. The file data.csv is the main dataset used for all Random Forest, XGBoost, and DML analyses. The files PPS_trend.csv and spot_hour_2024.csv are used to generate Figures 1, 2, and 4. The Python scripts used to produce the summary statistics, machine learning results, and figures are also included. Because all DML analyses and summary statistics are based on the results of Random Forest and XGBoost, please first run figureA1.1.py and figureA1.2.py, and then run the other Python files. Before running the Python scripts, please change the file paths to match your local directory.

用于生成清洗后逗号分隔值（CSV）数据集的数据源已在上文列明。本文所使用的最终清洗后数据集以CSV格式提供。data.csv文件是开展所有随机森林（Random Forest）、XGBoost及双重机器学习（DML）分析的核心数据集。PPS_trend.csv与spot_hour_2024.csv文件用于生成图1、图2与图4。用于生成描述性统计量、机器学习实验结果及各类图表的Python脚本亦随附提供。鉴于所有双重机器学习（DML）分析与描述性统计均基于随机森林和XGBoost的输出结果，请先运行figureA1.1.py与figureA1.2.py，再执行其余Python脚本。在运行Python脚本前，请修改文件路径以匹配您的本地目录结构。