Scholarlink

Accurate financial asset return prediction is critical for risk management and portfolio optimization. This study employs Long Short-Term Memory (LSTM) networks to forecast asset returns while integrating risk measures such as Value at Risk (VaR) and Conditional Value at Risk (CVaR). Unlike traditional statistical models, the LSTM approach efficiently captures complex temporal dependencies in financial data, enhancing decision-making in volatile markets. A comparative analysis with models like GARCH and ARIMA demonstrates LSTM’s superior accuracy, evaluated using Mean Absolute Error (MAE) and Mean Squared Error (MSE). The findings underscore LSTM’s potential in risk-aware financial forecasting, paving the way for advanced machine learning applications in asset management.

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

LSTM-Based Prediction of Financial Asset Returns: A Deep Learning Approach for Risk Assessment and Portfolio Optimization

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • A. P. S. Pavan Kumar,
  • K. Snehith Reddy,
  • K. Poojitha,
  • Nirmal Keshari Swain,
  • M. Yugandhar,
  • G. V. Chandra Shekar

摘要

Accurate financial asset return prediction is critical for risk management and portfolio optimization. This study employs Long Short-Term Memory (LSTM) networks to forecast asset returns while integrating risk measures such as Value at Risk (VaR) and Conditional Value at Risk (CVaR). Unlike traditional statistical models, the LSTM approach efficiently captures complex temporal dependencies in financial data, enhancing decision-making in volatile markets. A comparative analysis with models like GARCH and ARIMA demonstrates LSTM’s superior accuracy, evaluated using Mean Absolute Error (MAE) and Mean Squared Error (MSE). The findings underscore LSTM’s potential in risk-aware financial forecasting, paving the way for advanced machine learning applications in asset management.