What does this research mean for the field?

Integrating artificial neural networks with feature selection and price denoising achieves 70% to 85% forecasting accuracy for Indonesian equities, revealing that short-term momentum and value factors are the dominant predictors, particularly for Islamic equities. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To assess the predictive efficacy of combining deep learning models, feature selection techniques, and price denoising within an asset pricing framework for emerging market conventional and Islamic equities.

September 10, 2026International Journal of Islamic and Middle Eastern Finance and ManagementOpen Access

Deep learning in factor investing: an application to Indonesian Islamic and conventional equities

View Full Paper

Ask AI

Bookmark

Authors

MSMuhammad Ghazali Ash Shiddiqie MAMohsin Ali

Discussion

Member takes

Overview

Quantitative analysis demonstrates artificial neural networks predict equity returns across Indonesian stocks, indicating factor risk premia remain robust without deep architectural complexity.

Key Points

To assess the predictive efficacy of combining deep learning models, feature selection techniques, and price denoising within an asset pricing framework for emerging market conventional and Islamic equities.
Analyzed momentum, value, and quality risk premia across 949 conventional and 621 Islamic equities in Indonesia between 2016 and 2025.
Integrated artificial neural networks (ANN) with sequential feature selection (SeFS) and LASSO regularization.
Applied complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) for stock price denoising.
Artificial neural networks achieved strong predictive performance, with accuracy metrics ranging from 70% to 85%.
One-month momentum, earnings-to-price, and gross profit-to-total assets emerged as primary predictors, with Islamic stocks displaying higher sensitivity to valuation ratios such as EBIT/EV.
Predictive outcomes displayed stability across varying numbers of hidden layers, showing that deeper network architecture did not materially enhance forecast accuracy.

Cite This Study

Shiddiqie et al. (2026) studied this question.

synapsesocial.com/papers/6aa27b4758559d80afc7473e https://doi.org/10.1108/imefm-05-2026-0378

View Full Paper

Ask AI

Bookmark

September 10, 2026International Journal of Islamic and Middle Eastern Finance and ManagementOpen Access

Deep learning in factor investing: an application to Indonesian Islamic and conventional equities

View Full Paper

Ask AI

Bookmark

Authors

MSMuhammad Ghazali Ash Shiddiqie MAMohsin Ali

Discussion

Member takes

Overview

Quantitative analysis demonstrates artificial neural networks predict equity returns across Indonesian stocks, indicating factor risk premia remain robust without deep architectural complexity.

Key Points

To assess the predictive efficacy of combining deep learning models, feature selection techniques, and price denoising within an asset pricing framework for emerging market conventional and Islamic equities.
Analyzed momentum, value, and quality risk premia across 949 conventional and 621 Islamic equities in Indonesia between 2016 and 2025.
Integrated artificial neural networks (ANN) with sequential feature selection (SeFS) and LASSO regularization.
Applied complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) for stock price denoising.
Artificial neural networks achieved strong predictive performance, with accuracy metrics ranging from 70% to 85%.
One-month momentum, earnings-to-price, and gross profit-to-total assets emerged as primary predictors, with Islamic stocks displaying higher sensitivity to valuation ratios such as EBIT/EV.
Predictive outcomes displayed stability across varying numbers of hidden layers, showing that deeper network architecture did not materially enhance forecast accuracy.

Cite This Study

Shiddiqie et al. (2026) studied this question.

synapsesocial.com/papers/6aa27b4758559d80afc7473e https://doi.org/10.1108/imefm-05-2026-0378

Also Consider

Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark