Cointegration and Causality Relationship between Renewable Energy Indices

Authors

  • Fazlı IRMAK Ondokuz Mayıs Üniversitesi, Bafra Meslek Yüksekokulu, Muhasebe ve Vergi Bölümü, Çanakkale, Türkiye

DOI:

https://doi.org/10.20491/isarder.2026.2181

Keywords:

Renewable energy, PCA, Cointegration, VECM, Impulse–response function

Abstract

Purpose – The aim of this study is to examine the short- and long-run relationships between the renewable energy sector index constructed for Türkiye and selected renewable energy sectoral indices traded in the U.S. and global markets. Within this scope, the study evaluates the extent to which Türkiye’s renewable energy market is integrated with international renewable energy indices.
Design/methodology/approach – The analysis employs daily closing prices for the period 02.01.2020–10.01.2025. A Türkiye-specific renewable energy index (YES) was constructed using the PCA method based on the stock prices of 24 renewable energy companies listed on Borsa Istanbul. After confirming that all series are integrated of order one, I (1), through ADF, PP, and Zivot–Andrews unit root tests, we examine long-run relationships using the Gregory–Hansen cointegration test with structural breaks. Short- and long-run dynamics are analyzed using a Vector Error Correction Model (VECM). The direction of causality is assessed using Granger causality tests, while impulse–response analysis is employed to evaluate how U.S. and global renewable energy indices respond to shocks originating from the YES index.
Results – The results indicate the existence of a long-run cointegration relationship between the YES index and the US/global renewable energy indices in the presence of structural breaks. According to the long-run VECM coefficients, the YES index is positively associated with MAC and GSPTXCT but negatively related to GWE, RENIXX, ERIX, and DWCAEG. The error correction coefficient is estimated as -0.009458, implying that deviations from the long-run equilibrium in the YES index are corrected in approximately 106 days. In the short run, fluctuations in the YES index are primarily driven by GWE, RENIXX, GSPTXCT and DWCAEG. Granger causality analysis reveals bidirectional causality between YES and GWE and unidirectional causality running from YES to MAC, DWCAEG, and RENIXX.
Discussion – The findings demonstrate that Türkiye’s renewable energy market is significantly linked to international renewable energy indices in both the short and long runs, indicating a strong degree of market integration. Such integration implies that pricing dynamics in global renewable energy markets exert a substantial influence on Türkiye’s domestic market. The results further suggest that diversification opportunities within the renewable energy sector are limited, whereas the potential for spillover and contagion risk remains high, which is an important consideration for portfolio managers and investors. The construction of the YES index as a Türkiye-specific benchmark provides a more accurate basis for evaluating the integration of domestic renewable energy markets with their global counterparts and enhances the analytical clarity of cross-market linkages.

Published

2026-03-21

How to Cite

IRMAK, F. (2026). Cointegration and Causality Relationship between Renewable Energy Indices. Journal of Business Research - Turk, 18(1), 286–309. https://doi.org/10.20491/isarder.2026.2181

Issue

Section

Articles