Combining Different Technical Indicators for Sector Specific Stock Portfolio

Abstract

Financial markets data have non-straight, dynamic, loud, and non-organized data, in this way forecast the cost and discovered a trend are very challenging tasks. Among the most recent methods, machine learning procedures are with high profitability in the expectation of financial exchange forecast. Some of the widely used machine learning techniques in the financial market forecast are Support Vector Machine (SVM) and Neural Networks (RNN, CNN, etc.) apart from SVM and NN, logistic regression, Decision trees, and random forest are also used to predict the financial market prediction. it was concluded that research on the financial market is still relevant and that the use of data from stock markets is a highly researched topic [1]. Technical market indicators are the most common and widely used technique for predicting the cost of the equity market based on different periods by dissecting measurable patterns assembled from exchanging action, for instance, price movement in exchange and volume. Fundamental analysis is a strategy for estimating an equity's inborn stimulant looking to related monetary factors, such as development rate, Dividend Discount Model, Cash flow, etc. The goal of this study is to compare both CNN & RNN using different technical indicators for a specific stock portfolio. In this paper, we propose a CNN and RNN with a specific ordered feature set to predict the specific stock portfolio. The feature set is utilized using different technical indicators with alternative time intervals for both CNN and RNN.

Keywords: Financial market forecasting, Machine Learning, Support Vector Machine (SVM), Portfolio, Neural Networks (CNN, RNN), Technical Indicators, Fundamental Information.