Forecasting Bitcoin Price using Artificial Neural Network

doi:10.2139/ssrn.3515702

Paper

Forecasting Bitcoin Price using Artificial Neural Network

Published Jan 8, 2020 · Dhruba Banjade

Capital Markets: Asset Pricing & Valuation eJournal

UNKNOWN SJR score

2

Citations

0

Influential Citations

Full text

Semantic Scholar

Abstract

Forecasting Bitcoin prices is a big challenge today because of difficulties to exactly know the parameters upon which the Bitcoin price depends. In this paper, I forecast the price using the artificial neural network (ANN) technique. Bitcoin market capitalization, volume, dollar to euro, dollar to pound exchange rates, gold price, and S&P 500 index does not support to predict the prices. However, the Bitcoin price lags with four inputs, ten hidden layers, and two hundred iterations provide a better result with root mean square (RMSE) equals to 10.20, and coefficient of determination (R2) equals to 0.96.

Study Snapshot

Key takeawayThe artificial neural network technique using four inputs, ten hidden layers, and 200 iterations effectively forecasts Bitcoin prices with a root mean square of 10.20 and coefficient of determination of 0.96.

PopulationOlder adults (50-71 years)

Sample size24

MethodsObservational

OutcomesBody Mass Index projections

ResultsSocial networks mitigate obesity in older groups.

Sign up to use Study Snapshot

Consensus is limited without an account. Create an account or sign in to get more searches and use the Study Snapshot.

Create an account

Paper

Forecasting Bitcoin Price using Artificial Neural Network

Sign up to use Study Snapshot

References

Bitcoin price forecasting with neuro-fuzzy techniques

The PATSOS model, a hybrid Neuro-Fuzzy controller, effectively forecasts Bitcoin's daily price changes and provides higher investment returns compared to a naive buy-and-hold strategy.

Quantification of energy and carbon costs for mining cryptocurrencies

Cryptomining consumes more energy than conventional mining to produce one US$ of digital assets, with the exception of aluminium, and is responsible for 3-15 million tonnes of CO2 emissions.

Testing for asymmetric nonlinear short- and long-run relationships between bitcoin, aggregate commodity and gold prices

Bitcoin price movements can be predicted using aggregate commodity index and gold prices, but the relationships between Bitcoin and these asset classes are often asymmetric, nonlinear, and quantiles-dependent.

Bitcoin is not the New Gold – A comparison of volatility, correlation, and portfolio performance

Bitcoin and Gold have fundamentally different properties as assets and linkages to equity markets, with Bitcoin positively correlating with downward markets and no evidence of stable hedging capabilities.

Factors Influencing Cryptocurrency Prices: Evidence from Bitcoin, Ethereum, Dash, Litcoin, and Monero

Cryptomarket-related factors like market beta, trading volume, and volatility significantly influence cryptocurrency prices, while attractiveness matters only in the long run.

Citations

The Neuromarketing Concept in Artificial Neural Networks: A Case of Forecasting and Simulation from the Advertising Industry

Artificial neural networks (AI) are a cost-effective and powerful alternative to traditional neuromarketing tools for predicting consumer buying behavior in advertising campaigns.

A Stock Price Prediction Model Based on Investor Sentiment and Optimized Deep Learning

The MS-SSA-LSTM model, which combines sentiment analysis, swarm intelligence algorithm, and deep learning, outperforms other models in stock price prediction and is highly applicable.