Özet:

Purpose - The purpose of this research is to study the model of Bitcoin, which has the largest capitalization in the cryptocurrency market and the largest volume of transactions, which best explains its volatility. Method - In the study; GARCH (Non-Natural Generalized Otoregressive Conditional Heteroskedastisity) class; GARCH, EGARCH, IGARCH, GJR, FIGARCH-BBM, FIGARCH-CHUNG, FIEGARCH, FIAPARCH-BBM, FIAPARCH-CHUNG and HYGARCH models were used. The study used the daily closing price values of Bitcoin from the 2860-day US Dollar (USD) between 30.04.2013 and 26.02.2021 as a set of data. The findings - The research finds that the best forecasts for Bitcoin are made with the HYGARCH model. The GARCH model is also the best forecast and highest performance for non-sample play; the FIAPARCH-BBM model for 1 day forecast and the FIGARCH-CHUNG model for 5 and 10 days forecast. Bitcoin price gambling is an important input in the pricing formulas, portfolio selection and risk measurement. Therefore, the modeling and prediction of the volatility in Bitcoin prices is of great importance for both individual and corporate investors and theorists. It is therefore believed that the research results will be beneficial for individual and corporate investors in portfolio risk predictions. In future research, the study of the performance of different volatility forecast models for different cryptocurrencies is also believed to provide new and useful information to individual and corporate investors.

Anahtar Kelimeler: