Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Volatility Forecasting An Empirical Study on Bitcoin Using Garch and Stochastic Volatility models

Hultman, Hugo LU (2018) NEKP01 20181
Department of Economics
Abstract (Swedish)
Cryptocurrencies are on the rise, with new financial assets, new frameworks
need to be developed. This thesis sets out to the examine the GARCH(1,1),
the bivariate-BEKK(1,1), and the Standard stochastic volatility model’s volatility
forecasting performance on BTC/USD, where the bivariate model is estimated
on both BTC/USD and ETH/USD closing price data. Furthermore,
three loss functions are used to evaluate the forecast accuracy for each model.
The functions are estimated using realized volatility based on BTC/USD data
on a minute per minute basis. The result indicates that the GARCH(1,1) is
the model that performs best regarding forecast accuracy. All three loss functions
rank the models accordingly; first the GARCH(1,1), second... (More)
Cryptocurrencies are on the rise, with new financial assets, new frameworks
need to be developed. This thesis sets out to the examine the GARCH(1,1),
the bivariate-BEKK(1,1), and the Standard stochastic volatility model’s volatility
forecasting performance on BTC/USD, where the bivariate model is estimated
on both BTC/USD and ETH/USD closing price data. Furthermore,
three loss functions are used to evaluate the forecast accuracy for each model.
The functions are estimated using realized volatility based on BTC/USD data
on a minute per minute basis. The result indicates that the GARCH(1,1) is
the model that performs best regarding forecast accuracy. All three loss functions
rank the models accordingly; first the GARCH(1,1), second the bivariate-
BEKK(1,1), and finally the Stochastic volatility model. (Less)
Please use this url to cite or link to this publication:
author
Hultman, Hugo LU
supervisor
organization
course
NEKP01 20181
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Volatility forecasting, Stochastic volatility, GARCH, Bitcoin
language
English
id
8958504
date added to LUP
2018-09-26 09:13:39
date last changed
2018-09-26 09:13:39
@misc{8958504,
  abstract     = {{Cryptocurrencies are on the rise, with new financial assets, new frameworks
need to be developed. This thesis sets out to the examine the GARCH(1,1),
the bivariate-BEKK(1,1), and the Standard stochastic volatility model’s volatility
forecasting performance on BTC/USD, where the bivariate model is estimated
on both BTC/USD and ETH/USD closing price data. Furthermore,
three loss functions are used to evaluate the forecast accuracy for each model.
The functions are estimated using realized volatility based on BTC/USD data
on a minute per minute basis. The result indicates that the GARCH(1,1) is
the model that performs best regarding forecast accuracy. All three loss functions
rank the models accordingly; first the GARCH(1,1), second the bivariate-
BEKK(1,1), and finally the Stochastic volatility model.}},
  author       = {{Hultman, Hugo}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Volatility Forecasting An Empirical Study on Bitcoin Using Garch and Stochastic Volatility models}},
  year         = {{2018}},
}