An Analysis of Bitcoin Price Prediction Using Parametric Time-Series Forecasting Models

Résumé fourni par la source

Bitcoin is the most widely used cryptocurrency globally. Bitcoin is currently being invested in by many investors and regular people. Nevertheless, it becomes very challenging to assess or predict the price of Bitcoin. It is difficult to predict the future of digital money, a distinct kind of payment created using encryption techniques. Parametric models typically require fewer computational resources and are easier to implement than complex non-parametric or machine learning models. They can be quickly trained on standard software and are accessible to users without a deep understanding of machine learning. Prophet is a model for parametric time-series forecasting that uses additive decomposition. We used eight years' worth of historical Bitcoin data for our study, from 2012 to 2020. We compare the Fb-Prophet with the traditional Auto-Regressive Integrated Moving Average (ARIMA) and a deep learning model. The FB-Prophet works well compared to the other two models. Investors can use prediction models to determine when to purchase, hold, or sell Bitcoin. In an extremely volatile market, precise predictions can maximize profits and minimize losses by optimizing entry and exit locations.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.