×

You are using an outdated browser Internet Explorer. It does not support some functions of the site.

Recommend that you install one of the following browsers: Firefox, Opera or Chrome.

Contacts:

+7 961 270-60-01
ivdon3@bk.ru

Integrated usage of a linear regression model and a neural network in the problem of predicting the trend of Bitcoin cryptocurrency quotes

Abstract

Integrated usage of a linear regression model and a neural network in the problem of predicting the trend of Bitcoin cryptocurrency quotes

Alymova E.V.

Incoming article date: 10.10.2020

The paper presents an integrated approach using machine learning methods for choosing a trading strategy on the currency exchange. The presented approach uses the calculation of the linear regression angle coefficient by log return indicators and determination of the BTC/USD currency pair quotes trend in the next period based on the calculated coefficient sign. The feedforward multilayer neural network is used to predict the angle coefficient value in the next ten minute period for the current twenty minute period. The paper proposes a combined approach to the use of machine learning methods for choosing a trading strategy in foreign exchange. The study presents the results of experiments evaluating the practical results of effective and ineffective strategies based on the predicted values ​​of linear regression coefficients.

Keywords: diversification of management, production diversification, financial and economic purposes of a diversification, technological purposes of ensuring flexibility of production