A carregar...
Publicação
Estimating Value at Risk Assuming Pareto Tails: a Semiparametric Approach. Case Study for Cryptocurrencies
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.87 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
The main purpose of this master thesis is to determine the applicability of using Pareto tails to estimate the Value at Risk for cryptocurrencies. This study used five different cryptocurrencies, namely Bitcoin, Ether, Binance Coin, Ripple and Cardano. Concerning the methodology, six methods were used: Generalized Pareto Distribution, Normal Distribution, Historical Simulation, Gaussian GARCH, Student’s-t GARCH and Integrated GARCH in order to determine VaR for investment horizons of one week, one month and one quarter, each with two confidence intervals of 95% and 99%. The general timeframe considered was from 06/11/2017 to 06/11/2022, but a smaller timeframe has also been analyzed to study the impact of extreme events, such as the COVID-19 pandemic. The main takeaway from this research is that the semiparametric approach can indeed estimate the VaR of cryptocurrencies with high levels of precision in terms of marginal coverage, for both larger timeframes and extreme events.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Risk Analysis and Management
Palavras-chave
Value at Risk (VaR) Cryptocurrencies Pareto Tails GARCH Kupiec Christoffersen Covid-19 SDG 8 - Decent work and economic growth SDG 9 - Industry, innovation and infrastructure SDG 17 - Partnerships for the goals