Skip to content
All library documents

Forecasting Bitcoin Prices with Technical Indicators and Machine Learning

Article arXiv papers · Author: Samuel Asante Gyamerah
Read the original

Summary

This study evaluates whether machine learning models using technical indicators can forecast Bitcoin prices. It uses historical data from January 2012 to August 2019 and compares a penalized generalized linear model, random forest, linear support vector regression, and a stacking ensemble. The models are assessed with mean absolute percentage error, root mean square error, mean absolute error, and R-squared.

The reported best performer is a stack combining random forest and the generalized linear model, with linear support vector regression as its meta-learner. The document gives the stack’s error and fit metrics, which indicate a close match to prices in the study’s evaluation. However, it does not describe the validation setup, benchmark comparisons, or whether transaction costs and trading returns were considered. The results therefore concern price forecasting in the specified sample; they do not by themselves show that the method would generate profits or remain reliable in other periods.

Key ideas

  • The study uses technical indicators as predictors of Bitcoin prices.
  • It compares four machine learning approaches, including a stacking ensemble.
  • The selected ensemble combines random forest and a penalized generalized linear model beneath a support vector regression meta-learner.
  • Reported forecast metrics describe model fit but do not establish profitable trading performance.
  • The data cover January 2012 through August 2019.

Tags

CryptoMachine learningTechnical indicatorsStatisticsBacktesting
Full text
# Are Bitcoins price predictable? Evidence from machine learning techniques using technical indicators


# Are Bitcoins price predictable? Evidence from machine learning techniques using technical indicators









The uncertainties in future Bitcoin price make it difficult to accurately predict the price of Bitcoin. Accurately predicting the price for Bitcoin is therefore important for decision-making process of investors and market players in the cryptocurrency market. Using historical data from 01/01/2012 to 16/08/2019, machine learning techniques (Generalized linear model via penalized maximum likelihood, random forest, support vector regression with linear kernel, and stacking ensemble) were used to forecast the price of Bitcoin. The prediction models employed key and high dimensional technical indicators as the predictors. The performance of these techniques were evaluated using mean absolute percentage error (MAPE), root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R-squared). The performance metrics revealed that the stacking ensemble model with two base learner (random forest and generalized linear model via penalized maximum likelihood) and support vector regression with linear kernel as meta-learner was the optimal model for forecasting Bitcoin price. The MAPE, RMSE, MAE, and R-squared values for the stacking ensemble model were 0.0191%, 15.5331 USD, 124.5508 USD, and 0.9967 respectively. These values show a high degree of reliability in predicting the price of Bitcoin using the stacking ensemble model. Accurately predicting the future price of Bitcoin will yield significant returns for investors and market players in the cryptocurrency market.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.