Scholarlink

<p>This paper introduces a novel framework based on WT along with a DL model that incorporates GRU and CNN for option pricing in the Indian derivatives market. The goal of predicting real option prices is a crucial component in the financial markets since it determines trading mechanisms and risks in decision-making and financial development. The principal contribution of the paper is the wavelet transform allowing for a better understanding of the low-frequency and high-frequency option price time series movements. The research also introduces the use of GRU-CNN combines the temporal analysis from the GRU and the feature extraction from the CNN to capture the intricate decision-making in the option price data. By performing comprehensive experimental analysis on the real NIFTY50 option data the study concludes that the proposed WT-GRU-CNN model outperforms most of the purely deep learning-based models and other kinds of hybrid models. Using the above, the performances of the employed models are compared using a vast majority of the standard assessment indices used in the real world including RMSE, MAPE, MAE, and <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(R^2\)</EquationSource> </InlineEquation>. The findings point out that the WT-DL framework is better and more effective when compared to the conventionally used techniques in price prediction for options trading in the new-age Indian derivatives market. Thus, the study makes a contribution to the existing literature by demonstrating how signal processing and deep learning approaches can be effectively employed in the context of financial time series forecasting.</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

Leveraging Wavelet Transform & Deep Learning for Option Price Prediction: Insights from the Indian Derivative Market

  • 期刊论文

  • 发表日期：

    2025年9月23日

查看全文
  • Akanksha Sharma,
  • Chandan Kumar Verma

摘要

This paper introduces a novel framework based on WT along with a DL model that incorporates GRU and CNN for option pricing in the Indian derivatives market. The goal of predicting real option prices is a crucial component in the financial markets since it determines trading mechanisms and risks in decision-making and financial development. The principal contribution of the paper is the wavelet transform allowing for a better understanding of the low-frequency and high-frequency option price time series movements. The research also introduces the use of GRU-CNN combines the temporal analysis from the GRU and the feature extraction from the CNN to capture the intricate decision-making in the option price data. By performing comprehensive experimental analysis on the real NIFTY50 option data the study concludes that the proposed WT-GRU-CNN model outperforms most of the purely deep learning-based models and other kinds of hybrid models. Using the above, the performances of the employed models are compared using a vast majority of the standard assessment indices used in the real world including RMSE, MAPE, MAE, and \(R^2\) . The findings point out that the WT-DL framework is better and more effective when compared to the conventionally used techniques in price prediction for options trading in the new-age Indian derivatives market. Thus, the study makes a contribution to the existing literature by demonstrating how signal processing and deep learning approaches can be effectively employed in the context of financial time series forecasting.