Data · 17 Nov 2026
Machine Learning Prediction of Corporate Financial Performance Using ESG and Governance Data: An Incremental, Out-of-Sample Approach
Yasean Tahath-index 13, Yara Ibrahim, Ahmed Hassaneinh-index 34
Citations by database
Each database indexes a different set of journals, preprints and books, so they are shown side by side rather than blended.
- OpenCitations0
- Crossref0
Citations per year
The current year is still counting.
- 2026 so far0
Abstract
This study examines whether environmental, social, and governance (ESG) variables improve the prediction of one-period-ahead return on assets (ROA) conditional on current profitability, firm size, leverage, market valuation, industry, and country. The empirical analysis is based on 373 publicly listed firms across 16 European countries, yielding 4849 firm-period observations, of which 3968 have a measurable one-period-ahead ROA. We assess predictive performance using elastic net, random forest, extremely randomized trees, and histogram gradient boosting models across nested information sets. Model hyperparameters are selected using expanding-window validation, followed by a final two-period out-of-time evaluation. The extremely randomized trees specification incorporating financial, ESG-pillar, and governance variables produces the strongest predictive performance, with an out-of-sample mean absolute error (MAE) of 0.0315, root mean squared error (RMSE) of 0.0648, and R2 of 0.481. Relative to the financial-only specification, the expanded information set reduces MAE by 0.0003 and RMSE by 0.0004. The magnitude and statistical reliability of these gains, however, differ across predictive loss measures. Firm-cluster bootstrap estimates indicate that the governance-and-audit increment over the financial-plus-ESG model is statistically reliable, whereas the overall improvement over the financial-only model is positive but only marginally reliable; under a strictly held-out calibration period, the nominal 90% prediction interval achieves 81.4% empirical coverage, indicating imperfect calibration of predictive uncertainty. Historical ROA, firm size, industry, and country are the principal sources of predictive information, whereas audit-committee expertise and board gender diversity contribute less incremental information. The findings indicate that ESG and governance variables provide modest incremental improvements in predicting future ROA after conditioning on conventional financial information. Their predictive contribution is therefore better characterized as a refinement of forecast accuracy rather than a substitution for established financial predictors, with the magnitude of the incremental value depending on the choice of predictive model and evaluation criterion.
What the data shows
- No database records a citation yet, as is usual for a paper this new.
- Ahmed Hassanein has the highest h-index among the authors (34).
Graph
This paper in the middle, its authors and their most-cited papers below, the papers it cites to the left and the papers citing it to the right. The timeline lays every paper out by year.
Select a node for its details. Drag to pan, scroll to zoom, drag a node to move it.
Start with
Cited by 0 papers
Who cites this paper, from open citation data, with Crossref's record of each; the 100 newest when there are more.
References (40)
What it cites, from open citation data, with Crossref's record of each.
Sign in to see 20 more references
A free account opens the rest of the citing papers and references on every paper’s page.