Orman Finansmanında Yeşil Tahviller: Çevresel ve Finansal Belirleyicilere Nicelik-Üzerine-Nicelik ve Makine Öğrenmesi Perspektifinden Bir Bakış

Green bonds (GBs) have emerged as a prominent sustainable finance instrument, yet their role in funding timber and forestry (TF) projects remains underexplored. This study provides the first empirical assessment of the financial significance of GBs within the TF sector, addressing critical questions about their alignment with forestry’s risk–return profile. Using daily data we analyze the dynamic and nonlinear interactions between timber and forestry returns and green bonds performance. We also include control variables that might have significant impact on timber and forestry return series. Our multi-stage methodology integrates Quantile Cointegration, Quantile-on-Quantile analysis, and Random Forest machine learning to capture distributional heterogeneity, asymmetric effects in financial time series. Our results show that the green bonds have only a modest and asymmetric impact on timber and forestry, which is becoming significant mainly at distributional extremes. Random Forest analysis identifies GB performance and water availability as key drivers of TF returns, while carbon markets exert more moderate influences. This limited role of green bonds reflects persistent sector risks, including long project durations, changing environmental conditions and a lack of standardized metrics. Our findings indicate that key control variables such as water availability and carbon credits have significant, widespread effects on TF returns. By bridging econometric and machine learning approaches, this research advances methodological innovation in green finance studies and offers actionable insights for policymakers, investors, and forest sector stakeholders seeking to align capital flows with ecological outcomes.