Scholarlink

<p>In recent years, short squeeze events, such as the GameStop case in early 2021, have gained prominence, highlighting the need for advanced analyses of such phenomena. While traditional econometric and neural network approaches have struggled with predictive accuracy, our study addresses these gaps by analyzing the GameStop short squeeze using high-frequency intraday market data. We propose a novel hybrid approach that integrates an Autoregressive Moving Average-Generalized Autoregressive Conditional Heteroscedasticity model with Neural Networks, as well as exploiting fractal dynamics to capture multiscale temporal dependencies and hierarchical patterns in financial markets. This fractal framework effectively addresses the nonlinear and chaotic dynamics of the financial markets. Our methods deliver high predictive accuracy, with the ARMA-GARCH-Quantum approach standing out. This method highlights its greater adaptability and accuracy, proving the benefits of integrating fractal principles into predictive modeling. By enhancing adaptability and precision, this study contributes valuable tools for market forecasting and risk management, aiding regulators and financial managers in monitoring and mitigating abnormal price movements that could distort markets or spark crises.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

High-Frequency Trading, Short Squeeze and ARMA-GARCH-Fractal Neural Networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月5日

查看全文
  • David Alaminos,
  • M. Belén Salas-Compás,
  • Estefanía Alaminos

摘要

In recent years, short squeeze events, such as the GameStop case in early 2021, have gained prominence, highlighting the need for advanced analyses of such phenomena. While traditional econometric and neural network approaches have struggled with predictive accuracy, our study addresses these gaps by analyzing the GameStop short squeeze using high-frequency intraday market data. We propose a novel hybrid approach that integrates an Autoregressive Moving Average-Generalized Autoregressive Conditional Heteroscedasticity model with Neural Networks, as well as exploiting fractal dynamics to capture multiscale temporal dependencies and hierarchical patterns in financial markets. This fractal framework effectively addresses the nonlinear and chaotic dynamics of the financial markets. Our methods deliver high predictive accuracy, with the ARMA-GARCH-Quantum approach standing out. This method highlights its greater adaptability and accuracy, proving the benefits of integrating fractal principles into predictive modeling. By enhancing adaptability and precision, this study contributes valuable tools for market forecasting and risk management, aiding regulators and financial managers in monitoring and mitigating abnormal price movements that could distort markets or spark crises.