Value Investing

AlphaX: An AI-Based Value Investing Strategy for the Brazilian Stock Market ArXiv ID: 2508.13429 “View on arXiv” Authors: Paulo André Lima de Castro Abstract Autonomous trading strategies have been a subject of research within the field of artificial intelligence (AI) for aconsiderable period. Various AI techniques have been explored to develop autonomous agents capable of trading financial assets. These approaches encompass traditional methods such as neural networks, fuzzy logic, and reinforcement learning, as well as more recent advancements, including deep neural networks and deep reinforcement learning. Many developers report success in creating strategies that exhibit strong performance during simulations using historical price data, a process commonly referred to as backtesting. However, when these strategies are deployed in real markets, their performance often deteriorates, particularly in terms of risk-adjusted returns. In this study, we propose an AI-based strategy inspired by a classical investment paradigm: Value Investing. Financial AI models are highly susceptible to lookahead bias and other forms of bias that can significantly inflate performance in backtesting compared to live trading conditions. To address this issue, we conducted a series of computational simulations while controlling for these biases, thereby reducing the risk of overfitting. Our results indicate that the proposed approach outperforms major Brazilian market benchmarks. Moreover, the strategy, named AlphaX, demonstrated superior performance relative to widely used technical indicators such as the Relative Strength Index (RSI) and Money Flow Index (MFI), with statistically significant results. Finally, we discuss several open challenges and highlight emerging technologies in qualitative analysis that may contribute to the development of a comprehensive AI-based Value Investing framework in the future ...

Reports of Value’s Death May Be Greatly Exaggerated ArXiv ID: ssrn-3488748 “View on arXiv” Authors: Unknown Abstract Value investing, as defined by the Fama–French HML factor, has underperformed growth investing since 2007, producing a drawdown of 55% as of mid-2020. The under Keywords: Value investing, HML factor, Underperformance, Drawdown, Equities Complexity vs Empirical Score Math Complexity: 3.5/10 Empirical Rigor: 7.0/10 Quadrant: Street Traders Why: The paper uses standard finance statistics (drawdowns, percentiles, factor decomposition) rather than advanced mathematics, but its arguments are heavily grounded in empirical data analysis (55% drawdown, capitalization of intangibles, 2.2% annual return improvement, FANMAG stock attribution) and historical backtesting. flowchart TD A["Research Goal<br>Why has value investing underperformed?"] --> B["Methodology<br>Long-short HML factor portfolio"] A --> C["Data Inputs<br>Fama-French HML factor<br>2007-2020 period"] B --> D["Computational Process<br>Calculate cumulative returns & drawdown"] C --> D D --> E["Key Finding<br>55% drawdown observed"] D --> F["Key Finding<br>Value underperformed growth"] E --> G["Outcome<br>Value's underperformance<br>is severely underestimated"] F --> G

Deep Value ArXiv ID: ssrn-3122327 “View on arXiv” Authors: Unknown Abstract We define “deep value” as episodes where the valuation spread between cheap and expensive securities is wide relative to its history. Examining deep v Keywords: Deep Value, Value Investing, Valuation Spreads, Asset Pricing Anomalies, Quantitative Equity, Equity Complexity vs Empirical Score Math Complexity: 2.0/10 Empirical Rigor: 7.0/10 Quadrant: Street Traders Why: The paper uses straightforward descriptive statistics and historical analysis of valuation spreads, with minimal advanced mathematics, but appears heavily reliant on real market data and backtesting scenarios for its conclusions. flowchart TD A["Research Goal: Identify & model "Deep Value" episodes<br>widest valuation spreads relative to history"] --> B["Data & Inputs"] B --> B1["Panel of US Stocks"] B --> B2["Valuation Metrics<br>e.g., B/M, E/P"] B --> B3["Historical Time Series<br>for spread distribution"] B --> B4["Market Cap & Returns"] B --> C["Key Methodology: Deep Value Definition"] C --> C1["Compute cross-sectional valuation spread<br>e.g., Value - Growth spread"] C --> C2["Define Deep Value episodes<br>periods where spread > 90th percentile of history"] C --> D["Computational Process: Portfolio Construction"] D --> D1["Sort stocks into Value quantiles"] D --> D2["Go long Cheapest (Deep Value) decile<br>Short Expensive decile"] D --> D3["Calculate factor returns & alphas<br>controlling for momentum/quality"] D --> E["Key Findings & Outcomes"] E --> E1["Deep Value spreads are cyclical & persistent<br>predicting long-term returns"] E --> E2["Value factor returns significantly higher<br>during Deep Value episodes"] E --> E3["Returns decay over short horizons<br>but rebound over 3-5 years"] E --> E4["Out-of-sample performance robust<br>across regions and time"]

Deep Value ArXiv ID: ssrn-3076181 “View on arXiv” Authors: Unknown Abstract We define “deep value” as episodes where the valuation spread between cheap and expensive securities is wide relative to its history. Examining deep value acros Keywords: Deep Value, Value Investing, Valuation Spreads, Asset Pricing Anomalies, Quantitative Equity, Equity Complexity vs Empirical Score Math Complexity: 5.0/10 Empirical Rigor: 6.5/10 Quadrant: Street Traders Why: The paper uses standard financial mathematics and Gordon’s growth model but is grounded in extensive empirical analysis across multiple asset classes with detailed data construction (522 value strategies, 3000 deep value episodes), backtesting, and statistical testing of competing theories. flowchart TD A["Research Goal: Define and analyze 'Deep Value' episodes"] --> B["Data Input: Historical valuation spreads<br>(e.g., Price-to-Book, Price-to-Earnings)"] B --> C["Computational Process:<br>Calculate z-scores of valuation spreads over time"] C --> D["Key Methodology:<br>Identify 'Deep Value' regimes when spread > threshold"] D --> E["Outcome: Deep Value portfolios<br>(Buy cheap, sell expensive)"] E --> F["Key Finding: Value spreads widen during crises,<br>offering premium when reverting"]

Fact, Fiction, and Value Investing ArXiv ID: ssrn-2595747 “View on arXiv” Authors: Unknown Abstract Value investing has been a part of the investment lexicon for at least the better part of a century. In particular the diversified systematic “value factor” or Keywords: Value Investing, Value Factor, Systematic Investing, Factor Investing, Equities Complexity vs Empirical Score Math Complexity: 1.5/10 Empirical Rigor: 8.0/10 Quadrant: Street Traders Why: The paper relies on accessible, industry-standard data for straightforward empirical tests, resulting in high empirical rigor, but uses minimal advanced mathematics or dense formulas, leading to low math complexity. flowchart TD A["Research Goal<br>Is the value factor robust<br>across time and geographies?"] --> B["Methodology<br>Longitudinal & cross-sectional analysis"] B --> C["Data Inputs<br>Global equities<br>Decades of historical data"] C --> D["Computational Process<br>Systematic value factor construction<br>Backtesting & attribution"] D --> E["Key Findings<br>Value factor persists but varies<br>Systematic implementation required"]