Scholarlink

<p>The increasing volatility and complexity of financial markets demand more reliable and accurate trading strategies capable of reducing false signals and improving decision-making. Addressing this need, this research proposes an advanced ensemble-based methodology, Ensemble Trading Indicators Analysis (ETIA), designed to enhance the precision and robustness of trading signals. ETIA integrates nine state-of-the-art moving average indicators including Simple Moving Average (SMA) Cross, Exponential Moving Average (EMA) &amp; MA Cross, Moving Average Convergence Divergence (MACD), Double EMA (DEMA), Triple EMA (TEMA), Adaptive Moving Average (AMA), Double Moving Average (DMA), and Weighted Moving Average (WMA) to overcome the inherent limitations of single-indicator approaches. The ensemble uses a weighted majority voting mechanism, assigning each indicator a signal value (+ 1, 0, or − 1), with weights optimized through historical performance and machine learning. Final decisions are categorized into five actionable classes: Strong Buy, Weak Buy, Hold, Weak Sell, and Strong Sell. To further enhance predictive capability, we incorporate a deep learning model tailored for complex time-dependent financial data and integrate asymmetric cryptography to ensure secure data transmission and algorithmic integrity. Experimental evaluations and extensive backtesting demonstrate the superiority of ETIA, achieving up to 37–65% improvement in signal accuracy, reducing false positives by 28%, and increasing average 30-day profit across selected stocks by 196–643&#xa0;K compared to traditional single-indicator methods. These results confirm that the proposed ensemble framework significantly improves trading performance, stability, and risk management across diverse market conditions.</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

ETIA: Ensemble trading indicator analysis for improved market forecasting with enhanced deep learning and asymmetric cryptography

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月27日

查看全文
  • Sangeeta Banik,
  • Saikat Maity,
  • Sanjay Chakraborty

摘要

The increasing volatility and complexity of financial markets demand more reliable and accurate trading strategies capable of reducing false signals and improving decision-making. Addressing this need, this research proposes an advanced ensemble-based methodology, Ensemble Trading Indicators Analysis (ETIA), designed to enhance the precision and robustness of trading signals. ETIA integrates nine state-of-the-art moving average indicators including Simple Moving Average (SMA) Cross, Exponential Moving Average (EMA) & MA Cross, Moving Average Convergence Divergence (MACD), Double EMA (DEMA), Triple EMA (TEMA), Adaptive Moving Average (AMA), Double Moving Average (DMA), and Weighted Moving Average (WMA) to overcome the inherent limitations of single-indicator approaches. The ensemble uses a weighted majority voting mechanism, assigning each indicator a signal value (+ 1, 0, or − 1), with weights optimized through historical performance and machine learning. Final decisions are categorized into five actionable classes: Strong Buy, Weak Buy, Hold, Weak Sell, and Strong Sell. To further enhance predictive capability, we incorporate a deep learning model tailored for complex time-dependent financial data and integrate asymmetric cryptography to ensure secure data transmission and algorithmic integrity. Experimental evaluations and extensive backtesting demonstrate the superiority of ETIA, achieving up to 37–65% improvement in signal accuracy, reducing false positives by 28%, and increasing average 30-day profit across selected stocks by 196–643 K compared to traditional single-indicator methods. These results confirm that the proposed ensemble framework significantly improves trading performance, stability, and risk management across diverse market conditions.