---
title: ML and Expected Returns: Predictive Power
url: https://www.ml-quant.com/papers/ssrn/4713982/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: SSRN 4713982
source_url: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4713982
featured: 2024-02-07
citations: unknown
topic: Asset Pricing & Factors
---


# ML and Expected Returns: Predictive Power

Predictive Power: Machine learning models using option-based estimators are more effective than traditional models in predicting stock returns, especially for stocks with liquid options.

- Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4713982
- Identifier: SSRN 4713982
- Released: 2023-02-02
- First featured: Quant Letter No. 36 (2024-02-07): https://www.ml-quant.com/issues/2024-02-07/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Asset Pricing & Factors

## Related

- [Correlated Demand Shocks and Asset Pricing](https://www.ml-quant.com/papers/ssrn/4705500/): The correlated demand shocks from institutional investors can increase risk in asset pricing, with stocks exposed to higher correlated demand showing higher market betas and risk premiums.
- [No Sparsity in Asset Pricing: Evidence from a Generic Statistical Test](https://www.ml-quant.com/papers/ssrn/4730259/): The paper introduces a statistical test to identify sparsity in high-dimensional factor models, concluding that less than ten factors can explain stock returns and dense models perform better than sparse ones.
- [The Expected Returns on Machine-Learning Strategies](https://www.ml-quant.com/papers/ssrn/4702406/): Despite high turnover rates and the selection of hard-to-arbitrage stocks, machine learning strategies can predict profitable returns that common risk factors cannot explain, according to a study.
- [In Search of the True Greenium](https://www.ml-quant.com/papers/ssrn/4744608/): The study introduces a robust green score and expected returns to calculate the greenium, the expected return of green securities compared to brown, which is found to be more negative in greener countries and over time.
- [Commodity Futures Characteristics and Asset Pricing Models](https://www.ml-quant.com/papers/ssrn/4746258/): The article shows that a latent-factor model using the Instrumented Principal Component Analysis methodology surpasses existing models in explaining variations in commodity futures returns, with momentum, expected shortfall, and idiosyncratic volatility as key factors.
- [Exploratory Control with Tsallis Entropy for Latent Factor Models](https://www.ml-quant.com/papers/arxiv/2211.07622/): The research uses Tsallis Entropy in models with latent factors to optimally control and explore the state space, proving that the optimal state distribution is q-Gaussian, which can be used in creating robust statistical arbitrage trading strategies.