Time Series Analysis for Bitcoin Close Price
Data from Yahoo Finance has been used to predict the daily close price of Bitcoin in USD.
Regression-based models, advanced exponential smoothing models and, autoregressive integrated moving average models (ARIMA) were utilized for this project. Additional variations of the regression and advanced exponential smoothing models were also constructed to ensure better forecasts.
Model evaluation was based on the MAPE and RMSE accuracy measures. With the daily data of Bitcoin prices vulnerable to sentiments of the crypto world and with the continual improvement in blockchain technology, short-term forecasting is well suited for time series forecasting of any cryptocurrency.