ANALYSIS OF FACTORS INFLUENCING THE PROBABILITY OF PURCHASING CARBON CREDIT INSURANCE BY COMPANIES LISTED ON THE STOCK EXCHANGE OF THAILAND

This study examines the financial and environmental factors associated with the probability that companies listed on the Stock Exchange of Thailand will purchase carbon credit insurance. The research is motivated by the increasing importance of climate risk management and carbon pricing mechanisms in the transition toward a low-carbon economy. Using firm-level data from 2022 to 2024, the study integrates statistical and machine learning approaches to analyze > corporate decision-making under environmental and financial uncertainty. Logistic regression is employed to estimate the association between explanatory variables and the likelihood of purchasing carbon credit insurance, while a random forest model is used to evaluate predictive performance and the relative importance of the explanatory variables. The independent variables include financial indicators, environmental indicators, and industry characteristics. The results indicate that environmental variables, particularly greenhouse gas emissions per employee and carbon intensity relative to firm assets, have high relative importance in predicting the purchase decision. The random forest model achieved an accuracy of 75% and an AUC of 0.8466, with a mean cross-validation AUC of 0.8681 \((\mathrm{SD}=0.0453)\), whereas the logistic regression model achieved an accuracy of 61% and an AUC of 0.7007. The findings provide empirical evidence relevant to corporate risk management and the development of carbon-related financial instruments in Thailand.