Predicting Time Series Values with LSTM Different Scenarios for Prediction

Stock price prediction is a key area of research within financial circles. With stock prices being notoriously volatile-, predicting them has always been a subject of study in technical analysis and a known challenge. Even financial analysts struggle to identify and predict winning stocks. This is where pattern recognition modeling comes to play a major role. As trending in the stock market is widely targeted by Machine Learning, this paper shows a comparative study of different known approaches and introduces a new one by combining some of the known ones. Our hypothesis is tested using various forms of traditional Machine Learning algorithms as well as an LSTM (long short-term memory) deep learning model.

Paper

Full text

PDF

Predicting Time Series Values with LSTM Different Scenarios for Prediction

Semantic Scholar · Computer Science · 2022

Abstract

Stock price prediction is a key area of research within financial circles. With stock prices being notoriously volatile-, predicting them has always been a subject of study in technical analysis and a known challenge. Even financial analysts struggle to identify and predict winning stocks. This is where pattern recognition modeling comes to play a major role. As trending in the stock market is widely targeted by Machine Learning, this paper shows a comparative study of different known approaches and introduces a new one by combining some of the known ones. Our hypothesis is tested using various forms of traditional Machine Learning algorithms as well as an LSTM (long short-term memory) deep learning model.

Similar papers

© 2026 NYSGPT2525 LLC