Clean energy stock price forecasting and response to macroeconomic variables: A novel framework using Facebook's Prophet, NeuralProphet and explainable AI.

The clean energy market analysis has seen a strong surge in the post-Paris agreement owing to its undeniable environmental sustainability. The present work investigates the predictability of clean energy investment in the US market by selecting eight sectoral stock indices. The predictive exercise is carried out separately during pre-COVID-19 and COVID-19 timelines to draw key behavioral aspects of the underlying sectors. After scrutiny of previous research, we identify a set of technical and macroeconomic variables as explanatory constructs. We then utilize Facebook's Prophet and NeuralProphet to find future figures for the clean energy indices. Finally, Explainable Artificial Intelligence (XAI) is used to draw deeper insights into the contribution patterns of the constituent explanatory variables and obtain their relative importance. The findings suggest that the future movements of the sectoral clean energy assets can be predicted with a very high level of accuracy. Also, the predictability marginally improves during the COVID-19 pandemic despite the unprecedented uncertainty. Technical indicators appear to be the dominant features, while market sentiment and fear exert significant influence. • Understanding the dynamics of US clean energy sector using predictive analysis • Modeling clean energy sectoral dependence in normal and COVID-19 regimes • Predictability is gauged using Facebook's Prophet and NeuralProphet methods. • Explainable AI tools are used to uncover deeper dependence structure. • Technical indicators, market sentiment and fear exert significant influence. [ABSTRACT FROM AUTHOR]

Copyright of Technological Forecasting & Social Change is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)