Overreaction as an indicator for momentum in algorithmic trading: A Case of AAPL stocks

Paper: arXiv 2602.18912

Authors: Szymon Lis, Robert Ślepaczuk, Paweł Sakowski

Abstract

This paper investigates whether short-term market overreactions can be systematically predicted and monetized as momentum signals using high-frequency emotional information and modern machine learning methods. Focusing on Apple Inc. (AAPL), we construct a comprehensive intraday dataset that combines volatility normalized returns with transformer-based emotion features extracted from Twitter messages. Overreactions are defined as extreme return realizations relative to contemporaneous volatility and transaction costs and are modeled as a three-class prediction problem. We evaluate the performance of several nonlinear classifiers, including XGBoost, Random Forests, Deep Neural Networks, and Bidirectional LSTMs, across multiple intraday frequencies (1, 5, 10, and 15 minute data). Model outputs are translated into trading strategies and assessed using risk-adjusted performance measures and formal statistical tests. The results show that machine learning models significantly outperform benchmark overreaction rules at ultra short horizons, while classical behavioral momentum effects dominate at intermediate frequencies, particularly around 10 minutes. Explainability analysis based on SHAP reveals that volatility and negative emotions, especially fear and sadness, play a central role in driving predicted overreactions. Overall, the findings demonstrate that emotion-driven overreactions contain a predictable structure that can be exploited by machine learning models, offering new insights into the behavioral origins of intraday momentum and the interaction between sentiment, volatility, and algorithmic trading.

Complexity vs Empirical Score

• Math Complexity: 3.5/10
• Empirical Rigor: 7.5/10
• Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper presents a solid empirical pipeline evaluating high-frequency Twitter emotion features combined with machine learning models (XGBoost, BiLSTM) to forecast intraday market overreactions in AAPL. While mathematically straightforward, relying primarily on standard ML classifiers and volatility-normalized returns, the empirical setup is rigorous across multiple intraday frequencies and includes transaction costs and SHAP explainability analysis.

Research Flowchart

  flowchart TD
    A[Research Goal: Predict & monetize intraday AAPL overreactions via ML & Twitter sentiment] --> B[Data Collection: High-Frequency AAPL Returns & Twitter Sentiment]
    B --> C[Feature Engineering: Volatility Normalization & Transformer-based Emotion Extraction]
    C --> D[Multi-Class ML Modeling: XGBoost, RF, DNN, BiLSTM across 1-15 min horizons]
    D --> E[Model Evaluation: Backtesting Trading Strategies vs. Benchmark Rules]
    E --> F[Explainability Analysis: SHAP Feature Importance]
    F --> G[Outcomes: ML outperforms at 1-min; classical momentum dominates at 10-min; fear/sadness drive overreaction]

Cited References (44)

1. Risk as feelings. (2001)
2. Performance hypothesis testing with the Sharpe ratio (2003)
3. Investor psychology and security market under-and overreactions (1998)
4. Fear and greed in financial markets: A clinical study of day-traders (2005)
5. Dynamics of stock market developments, financial behavior, and emotions (2023)
6. Emotion regulation and behavior in an individual decision trading experiment: Insights from psychophysiology (2015)
7. Thinking, feeling and deciding: The influence of emotions on the decision making and performance of traders (2011)
8. Emotions in the stock market (2020)

0 of 44 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.