Research Article
Prediction of Investment Revolution of Green Enterprises: A Two-dimension Approach to the Combination of Characteristics and Temporalities of Green-specific Characteristics
@INPROCEEDINGS{10.4108/eai.22-5-2026.2365224, author={Shiyao Lu}, title={Prediction of Investment Revolution of Green Enterprises: A Two-dimension Approach to the Combination of Characteristics and Temporalities of Green-specific Characteristics}, proceedings={Proceedings of the 4th International Conference on Image, Algorithms, and Artificial Intelligence, ICIAAI 2026, 22-24 May 2026, Singapore, Singapore}, publisher={EAI}, proceedings_a={ICIAAI}, year={2026}, month={8}, keywords={Green enterprises; Investment return prediction; ESG metrics; Temporal dependencies; Gradient boosting}, doi={10.4108/eai.22-5-2026.2365224} }- Shiyao Lu
Year: 2026
Prediction of Investment Revolution of Green Enterprises: A Two-dimension Approach to the Combination of Characteristics and Temporalities of Green-specific Characteristics
ICIAAI
EAI
DOI: 10.4108/eai.22-5-2026.2365224
Abstract
The paper suggests a dual-dimension feature system combining financial indicators, ESG measures, Personalized green features and lagged return features. The paper evaluates the regression algorithms on the use of a panel dataset of 300 observations that consist of 50 green enterprises (2018-2023). It is shown through the results of the experiment that Gradient Boosting algorithm provides optimal results, with a Mean Squared Error (MSE) of 0.011 and R-squared (R2) of 0.80. According to the feature importance analysis, the core predictors are the composite ESG rating, the ratio of green R&D investment, and the lagged returns. The proposed methodology is a powerful analytical tool that will enable investors in green finance to analyze and decide by quantifying green features and providing a combination of the static and dynamic dimensions.
