Explore the corpus

Find evidence, not just documents.

Search Minerva’s current Findings using transparent hybrid retrieval. Results remain strictly inside the configured corpus.

Hybrid retrieval

Results for “Machine Learning in Finance”

30 findings

72match

Finding 23941 source

Employing ensemble learning, SHAP values, and double machine learning, the research boosts performance while noting challenges like data manipulation, fake citations, and quality compromises, urging human-machine collaboration and market impact.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

71match

Finding 53131 source

A new study explains why highly complex machine learning models, with thousands of parameters, can excel in finance even with tiny datasets, challenging classical theory and resolving a longstanding paradox in the field.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

71match

Finding 23041 source

The study concludes that AI techniques, especially deep reinforcement learning, hold great potential for stock market prediction. Future research should focus on improving data availability, model interpretability, and real-time trading implementations. Potential applications include automated trading systems and enhanced risk management strategies.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

70match

Finding 53161 source

This paper uniquely resolves a major paradox in machine learning for finance by introducing 'projection learning,' a novel framework with exact finite-sample theory and closed-form expressions. It compellingly explains why highly complex models ($12{,}000$ parameters, $12$ observations) succeed, reconciling empirical results with theory and offering significant practical and theoretical impact.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

69match

Finding 53141 source

The paper introduces 'projection learning,' showing these models succeed by projecting the return function onto observable features, not by learning the true functionimproving risk-adjusted returns (Sharpe ratios) as complexity increases.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

69match

Finding 55501 source

This paper introduces a unified, end-to-end machine learning framework for portfolio optimization, directly integrating investor preferences, constraints, and transaction costs into the learning objective. Its originality lies in explicit modeling of investor heterogeneity and real-world factors, demonstrating empirical superiority over traditional methods, making it compelling for quantitative finance advancement.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

69match

Finding 23951 source

The paper presents an original synthesis of established corporate finance methods with innovative applications of $AI$ and $machine learning$. It introduces fresh perspectives on analyzing unstructured data, causal inference, and related issues. Though primarily building on existing ideas, its thoughtful integration offers readers valuable insights and compelling practical relevance overall.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

69match

Finding 53151 source

Using advanced random matrix theory and kernel methods, the author provides exact formulas for bias, variance, and performance, enabling millisecond calculations, though results depend on assumptions like i.i.d. data and specific regularity conditions.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

68match

Finding 20121 source

The study credits superior stock selection, lower expenses, and smart trading cost management for these gains, and finds that blending machine learning with human expertise delivers consistent outperformance across various market conditions and investment styles.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

68match

Finding 20111 source

U.S. equity mutual funds are increasingly leveraging machine learning, with a new measureActive Machine Learning Based Trading (AMLT)revealing that top AMLT funds outperform bottom decile peers by 2.4%--3.0% annually, risk-adjusted.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

68match

Finding 23931 source

The chapter integrates deep learning with explainable AI to demystify models, combining text, audio, video, and numerical data for insights into investor sentiment, financial distress, portfolio construction, and corporate governance.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

68match

Finding 20131 source

AMLT offers a precise way to track machine learning adoption, surpassing older methods based on fund names or staff skills, but the research notes limited exploration of ML risks and a small sample of self-identified AI funds.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

67match

Finding 41621 source

A key insight is that investing is fundamentally a causal problem, not just forecasting; robust attribution of returns to risk factors under uncertainty is essential, challenging standard machine learning tools in finance.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

67match

Finding 23921 source

The study demonstrates how artificial intelligence transforms corporate finance by enhancing prediction accuracy, uncovering hidden patterns, and analyzing unstructured data with models like random forests, neural networks, and causal forests.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

67match

Finding 20141 source

This paper introduces the novel AMLT measure, quantifying mutual funds alignment with forward-looking machine learning signals using both numerical and textual data. Its comprehensive, holdings-based approach uniquely distinguishes ML adoption in investment strategy. Large-scale empirical analysis and performance decomposition make it a compelling, original, and significant contribution to investment management literature.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 64601 source

Advanced machine learning models, such as gradient boosted trees and neural networks, predict financial returns more accurately than linear models, with performance tied to market conditions and retail trading activity.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 46651 source

The research demonstrates that deep generative models are effective for financial applications, such as risk analysis and portfolio optimization. Future work could explore additional architectures and features. Potential applications include creating optimal portfolios and improving risk management strategies.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 55471 source

A new end-to-end machine learning framework revolutionizes portfolio management by directly linking asset return prediction with optimization, outperforming traditional models, especially under real-world trading restrictions and diverse investor risk preferences.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 38511 source

A comprehensive study integrates machine learning techniques, including Extreme Gradient Boosting and Naive Bayes, with the classic Ichimoku Cloud strategy, enhancing trading performance and reducing risk through lower maximum drawdowns.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 55381 source

A new book chapter spotlights the Zero-Crossing Asset Pricing Model (ZCAPM), a machine learning approach using the expectation-maximization (EM) algorithm, as a breakthrough in predicting stock market anomalies more accurately than traditional models.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 37831 source

While the approach is innovative and promising, the paper calls for more detailed empirical results and discussion of limitations, including model interpretability and robustness across varying market conditions.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 55411 source

This paper introduces ZCAPM, applying the expectation-maximization algorithm to estimate a latent variable for asset return sensitivity to market dispersiona novel machine learning approach in asset pricing. Its empirical validation, outperforming traditional multifactor models and supporting the efficient market hypothesis, marks a significant, compelling advancement, despite methodological extensions.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 41641 source

This paper offers a structured synthesis of challenges in applying machine learning and AI to finance, reframing familiar issuescausality, non-stationarity, low signal-to-noise ratio, small sample sizesas engineering problems. Its originality lies in advocating for AI tailored to investing, though it lacks novel algorithms or groundbreaking insights, making it moderately compelling.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

66match

Finding 37811 source

Researchers introduce learning-to-rank, a machine learning technique from information retrieval, to asset pricing, arguing it offers more practical and informative stock ranking than traditional regression-based predictions in finance.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

65match

Finding 55391 source

ZCAPM uniquely explains the high returns of hundreds of anomaly portfolios in out-of-sample tests, challenging the belief that markets are inefficient and casting doubt on the necessity of behavioral finance explanations.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

65match

Finding 53971 source

This paper introduces the novel Limits-to-Learning Gap (LLG), a universal, data-driven lower bound quantifying the gap between empirical model fit and true population benchmarks in financial machine learning. Its originality lies in addressing underestimation of predictability, offering practical corrections, and influencing both academic research and real-world financial ML applications.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

65match

Finding 59151 source

Findings: models succeed in niches but fail at structural breaks; reflexivity constrains design. Strengths: continuous learning, regime recognition, meta-learning, adversarial robustness. Limits: subjective scoring, scarce tests, underexplored markets, execution edges.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

65match

Finding 59141 source

Three contributions: bridge aligning finance and ML; a stress test of PatchTST, non-stationary transformers, neural HMMs, TFT, MoE, online/meta-learning, RL; and a validation blueprintregime tests, adaptation latency, costs/impact, buy-and-hold benchmark.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

65match

Finding 53951 source

The study introduces the Limits-to-Learning Gap (LLG), a universal lower bound that quantifies the gap between what models can learn and true predictability, offering a practical diagnostic tool for financial data analysis.

LexicalMeaningQualityGraph

Matched: finance, learning, machine

65match

Finding 55401 source

By uncovering hidden signals indicating stocks reactions to market return dispersion, ZCAPM supports the efficient markets hypothesis, though the chapter notes it does not deeply explore the models limitations or implications for varying market conditions.

LexicalMeaningQualityGraph

Matched: finance, learning, machine