Integration of the Political Events in the Fossil Fuels Equity Market: a PCA and Forecasting Approach

Romain Alfred () and Hamza Chergui
Additional contact information
Romain Alfred: SKAIZen Group
Hamza Chergui: SKAIZen Group

Post-Print from HAL

Abstract: In this paper, we propose a methodology to test the integration of political events from the GDELT (Global database of events, language and tone) event database in the fossil fuels equity market prices. Our methodology is based on an approach borrowed from the field of financial time series forecasting. To represent the market to be predicted, we use the PCA technique (principal components analysis) to construct an index statistically representative of our market of interest, based on an equity portfolio. Our results show that political trends calculated on the basis of the political events and geopolitical analysis are features that improve forecasting, compared with delayed mathematical transformations of the time series alone. In the calculation of political trends, we also propose a partition of the international system into geopolitical spheres. As we explain in the article, our methodology represents a first step towards a better quantification of the political risks applied to investment.

Keywords: GDELT project; financial time series forecasting; machine learning; fossil fuels market; dimension reduction; geopolitical sphere; political risk (search for similar items in EconPapers)
Date: 2024-06-06
Note: View the original document on HAL open archive server: https://hal.science/hal-04608659
References: View complete reference list from CitEc
Citations:

Published in QFFE 2024: Quantitative Finance and Financial Econometrics International Conference, Aix-Marseille School of Economics, Jun 2024, Marseille, France

Downloads: (external link)
https://hal.science/hal-04608659/document (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:journl:hal-04608659

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().