The paper focuses on the similarities between green bonds and conventional bonds. A feature of green bonds is lower yields compared to conventional bonds with the same risk. The purpose of this study is to examine the determinants of the yields of selected bonds present in the global financial market. Among the financial characteristics, ESG rating was included as a determinant (Zhang et al. 2021; Immel et al. 2021). Macroeconomic factors, such as the consumer price index (CPI) or GDP growth rate, would also affect the size of the greenium (Cavallo and Valenzuela 2010; Ivashkovskaya and Mikhaylova 2020; Nanayakkara and Colombage 2019). There is no single set of determinants in the literature, as different studies have shown different results for different markets and countries, as well as types of markets. This study will use machine learning methods (e.g. based on gradient enhancement over decision trees – CatBoost). The findings will lead to a better understanding of the green bond market for investors, researchers, regulators and potential issuers.