- Monday, Apr 6th, 2026
| TITLE | Implementation of Kernel Adaptive Filtering in Stock Prices |
|---|---|
| ABSTRACT | Stock price prediction is a complex problem due to the volatile, nonlinear, and dynamic nature of financial markets. This project applies Kernel Adaptive Filtering KAF along with machine learning techniques such as Support Vector Machines SVM and Random Forest to model hidden dependencies in stock price movements. I believe while these approaches enhance predictive accuracy, challenges such as computational complexity, over fitting, parameter sensitivity, data dependency, scalability, and limited generalization remain. To rectify these issues, several strategies are incorporated dimensionality reduction and GPU-based computation to reduce complexity, regularization and cross-validation to prevent over fitting, automated hyper parameter tuning for parameter stability, hybrid modeling with sentiment analysis to handle external influences, online learning for scalability, and transfer learning with ensemble methods to improve adaptability across markets. By integrating these solutions, the framework enhances both the efficiency and reliability of stock price forecasting, making it more practical for real-world financial decision-making. |
| AUTHOR | M Abhilash, K Aryan , J Madhusudhan , E Bhanu Prakash , T Vardhan, Gattu Prasad Student, Dept. of CSE, Sphoorthy Engineering College, Hyderabad, India Assistant Professor, Dept. of CSE, Sphoorthy Engineering College, Hyderabad, India |
| VOLUME | 13 |
| DOI | DOI: 10.15680/IJARETY.2026.1302029 |
| 29_Implementation of Kernel Adaptive Filtering in Stock Prices.pdf | |
| KEYWORDS | |
| References | [1]Spectral Eigenfunction Decomposition for Kernel Adaptive Filtering Kan Li & Jose C. Principe (2025) Proposes a new method (SPEED) to make KAF scalable and efficient in time-series prediction tasks. Link: https://arxiv.org/abs/2501.08989 [2]An Analytic Solution for Kernel Adaptive Filtering Benjamin Colburn et al. (2024) Provides a closed-form analytical framework for KAF, useful for nonlinear time-series prediction such as stock prices. Link: https://arxiv.org/abs/2402.03497 [3]Robust Kernel Adaptive Filtering for Nonlinear Time Series Prediction (2023) Discusses a variant of kernel adaptive filters optimized for noisy and nonlinear time-series scenarios. Link: DOI/Link: https://doi.org/10.1016/j.sigpro.2023.109090 |
Copyright © IJARETY 2023 All Rights Reserved.