Predicting volatility of bitcoin returns with ARCH, GARCH and EGARCH models

Abstract The ınvestment decisions of institutional and individual investors in financial markets are largely influenced by market uncertainty volatility the investment instruments. Thus, prediction volatilities prices returns instruments becomes imperative for successful investment. In this study we seek to identify best fit model that can predict return Bitcoin, which is high demand as an tool recent times. Using opening data weekly Bitcoin period 11.24.2013–03.22.2020, their logarithmic were calculated. stationarity properties series was tested applying ADF unit root test found be stationary. After reaching average equation ARMA (2.2), it whether there ARCH effect (2,2) model. As a result applied ARCH-LM test, reached residuals selected have effect. Volatility after detection has been tried with conditional variance models such (1), (2), (3), GARCH (1,1), (1,2), (1,3), (2,1), (2,2), EGARCH (1,1) (1,2). While obtained findings indicate direction according Akaike info criterion, does not test. our empirical highlight ample guide on appropriate modeling price information market.