Classification of Stock Price Using Fundamental and Technical Analysis

Abstract

Investors who want to make well-informed decisions face problems due to the constantly shifting nature of financial markets. Pattern identification becomes more difficult when analyzing high-dimensional data because quality patterns become more complex. Various machine learning approaches are frequently used in this complex situation to efficiently anticipate and categorize future data. However, customizing a computer for a particular portion of the dataset usually comes at a high cost. This study suggests a deep learning method to recognize various price patterns in the stock market by utilizing long short-term memory (LSTM) networks. LSTM networks are superior to conventional methods in capturing temporal dependencies and patterns in time-series data, as they may not require as much training cost for certain dataset segments. A thorough review of a company's data and economic indicators is part of the fundamental analysis component of the study. Technical analysis is also used to look at trading volumes, historical price movements, and different technical indicators. It aims to get over the limitations of traditional approaches by using LSTM model capabilities to find periodic relationships and patterns in stock market data. Keywords : LSTM, Stock Patterns, Machine Learning, Stock Prediction, Technical Analysis