Scholarlink

<p>The stock market's price is the most important pointer of an economy's expansion. Predicting the exact fluctuations of the stock price on the market is necessary to make a profit. However, accurate assessments cannot be performed due to the stock market's intricate and unpredictable behavior. Thus, robust recognition models are highly desired for investors' financial decision-making processes. In this research, a deep learning model is recommended to forecast the stock’s price in the marketplace. The required data are collected from the Netflix Stock Price Prediction, Stock market prediction dataset, and the technical indicators form the dataset including momentum close price, the moving average of close price, and median price are taken to find the real-time situation of the market. In order to distinguish the real market situation, the technical indicators are generated from the input data. The technical indicators are further passed to the optimal weight-based feature selection phase, which is performed using the Enhanced Good and Bad Group Optimization (EGB-GpO). The attained optimal features are passed to the Adaptive Temporal Convolutional Network with the Bayesian Learning technique (ATCN-BL) to detect the price of the stocks in the marketplace. The results showed that the Mean Error Percentage (MEP) of the recommended model is 3.35, which is comparatively lower than the 6.65, 7.17, 5.55, and 4.26, accomplished by the existing approaches like RNN, TCN, BL, and TCN-BL, respectively. The suggested model is practically applied in identifying the overvalued or undervalued stocks to safeguard the investment to enhance the financial goals of investors. The suggested model helps to determine the moving price of the stocks in the market to guide the investor in making appropriate decisions to enhance their outcome.</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

A Full-Fledged Stock Market Prediction Framework using Adaptive TCN with a Bayesian Learning Network via Enhanced Good and Bad Groups-Based Optimizer

  • 期刊论文

  • 发表日期：

    2025年6月23日

查看全文
  • Rakesh Roshan,
  • Krishna Kumar N,
  • Supraja Ballari,
  • C Antony,
  • Surya Kiran Chebrolu,
  • Om Prakash Rishi,
  • V Biksham,
  • Kumar Neeraj,
  • Vangapally Raju

摘要

The stock market's price is the most important pointer of an economy's expansion. Predicting the exact fluctuations of the stock price on the market is necessary to make a profit. However, accurate assessments cannot be performed due to the stock market's intricate and unpredictable behavior. Thus, robust recognition models are highly desired for investors' financial decision-making processes. In this research, a deep learning model is recommended to forecast the stock’s price in the marketplace. The required data are collected from the Netflix Stock Price Prediction, Stock market prediction dataset, and the technical indicators form the dataset including momentum close price, the moving average of close price, and median price are taken to find the real-time situation of the market. In order to distinguish the real market situation, the technical indicators are generated from the input data. The technical indicators are further passed to the optimal weight-based feature selection phase, which is performed using the Enhanced Good and Bad Group Optimization (EGB-GpO). The attained optimal features are passed to the Adaptive Temporal Convolutional Network with the Bayesian Learning technique (ATCN-BL) to detect the price of the stocks in the marketplace. The results showed that the Mean Error Percentage (MEP) of the recommended model is 3.35, which is comparatively lower than the 6.65, 7.17, 5.55, and 4.26, accomplished by the existing approaches like RNN, TCN, BL, and TCN-BL, respectively. The suggested model is practically applied in identifying the overvalued or undervalued stocks to safeguard the investment to enhance the financial goals of investors. The suggested model helps to determine the moving price of the stocks in the market to guide the investor in making appropriate decisions to enhance their outcome.