Market Trend Prediction and Analysis Using ICEEMDAN and Time Series Algorithms

Sanjana Iyer, P. Ranjana, Neha Kirubakaran, Vengatesh M.

2025

Abstract

Financial market prediction is a complex task due to the non-linearity and high volatility of stock prices. This paper presents a hybrid model leveraging the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) for decomposing stock prices and a Long Short-Term Memory (LSTM) network for predictive modelling. ICEEMDAN effectively extracts intrinsic mode functions (IMFs), capturing stock price trends and fluctuations, while LSTM learns temporal dependencies. A Streamlit-based interactive system visualizes past stock trends and forecasts future prices. The proposed model is tested on real-time stock datasets using Yahoo Finance (yfinance) data. Results demonstrate the superiority of ICEEMDAN-based LSTM over conventional models in predicting stock market trends with improved accuracy and robustness.

Paper Citation

in Harvard Style

Iyer S., Ranjana P., Kirubakaran N. and M. V. (2025). Market Trend Prediction and Analysis Using ICEEMDAN and Time Series Algorithms. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 773-781. DOI: 10.5220/0013920600004919

in Bibtex Style

@conference{icrdicct`2525,
author={Sanjana Iyer and P. Ranjana and Neha Kirubakaran and Vengatesh M.},
title={Market Trend Prediction and Analysis Using ICEEMDAN and Time Series Algorithms},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={773-781},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013920600004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Market Trend Prediction and Analysis Using ICEEMDAN and Time Series Algorithms
SN - 978-989-758-777-1
AU - Iyer S.
AU - Ranjana P.
AU - Kirubakaran N.
AU - M. V.
PY - 2025
SP - 773
EP - 781
DO - 10.5220/0013920600004919
PB - SciTePress