Adaptive Neuro-Fuzzy System to Predict Cryptocurrency Variation in Stable and Crisis Periods: Case of COVID-19

22 Pages Posted: 12 Sep 2022

See all articles by Hager Chlif

Hager Chlif

University of Sousse

Dalel Kanzari

University of Sousse

yosra ridha bensaid

University of Sfax

Abstract

Cryptocurrencies have received significant attention in recent years, particularly during the crisis by studying the impact of the COVID'19 outbreak on the efficiency of the cryptocurrency market. Accurately predicting the cryptocurrency's price variation is a challenge given their volatility, risk factors, lack of accurate fundamental values and sensitivity to many apparent factors such as supply and demand, investor and user sentiment, government regulation and media hype, and other indistinct hidden factors. Many research studies use econometric and machine learning models to predict cryptocurrency price changes, but almost all focus on homogeneous financial data during stable or unstable periods.This paper proposes ANFPC, an Adaptive Neuro-Fuzzy Prediction Cryptocurrency to predict the crypto-currencies price variations, during stable and crisis periods, based on heterogeneous financial and non-financial data. ANFPC is a generic decision support system that combines a fuzzy logic model and artificial neural network (ANFIS) to accurately treat approximative and heterogeneous independent data. To assess the performance of the proposed model, we choose the COVID crisis as a crisis period and the Vaccine period as a stable period (post-COVID). The experimental results prove that ANFPC has accurate predictions better than ANN and LSTM.  The proposed approach provides a suitable decision support system for diverse types of investors: individuals and institutions.

Keywords: cryptocurrency, Prediction, ANFIS, Fuzzy logic, COVID'19 crisis

Suggested Citation

Chlif, Hager and Kanzari, Dalel and bensaid, yosra ridha, Adaptive Neuro-Fuzzy System to Predict Cryptocurrency Variation in Stable and Crisis Periods: Case of COVID-19. Available at SSRN: https://ssrn.com/abstract=4209093 or http://dx.doi.org/10.2139/ssrn.4209093

Hager Chlif

University of Sousse ( email )

rue Abdelaziz el Behi
Sousse, 4000
Tunisia

Dalel Kanzari

University of Sousse ( email )

rue Abdelaziz el Behi
Sousse, 4000
Tunisia

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
83
Abstract Views
372
Rank
539,499
PlumX Metrics