Price Analysis and Risk Management in Cryptocurrency Market
Abstract
Cryptocurrency and its underlying technology, blockchain, are significantly changing the financial world. One of the critical characteristics of cryptocurrency is its substantial price fluctuations, which means investment risk. This dissertation explores the cryptocurrency market by analyzing
its price formation and risk management. By investigating the inter-market relationship among
multiple cryptocurrencies, it is found that cryptocurrency’s price affects each other in the long run and short run, and the competition in the market matters in price formation. I forecast bitcoin’s
volatility and examine the cryptocurrency index (CRIX) tail behavior to address the cryptocurrency
risk management issue. I adopt the Value at Risk and Expected Shortfall as risk measurements, which measure the risk exposure in the cryptocurrency market. The results of bitcoin volatility
forecasting provide evidence that the RNN outperforms GARCH and EWMA in average forecasting performance. However, it is less efficient in capturing the bitcoin market’s extreme events.
Moreover, the RNN shows poor performance in Value at Risk forecasting, indicating that it could
not work well as the econometric models in explaining extreme volatility. By investigating the cryptocurrency index tail behavior, I discover that focusing on tail events using extreme value theory and filtered historical simulation methods yields more accurate Value at Risk and Expected
Shortfall estimation. This in turn significantly improves the efficiency of cryptocurrency index risk management. This dissertation is motivated by the fact that there is an increasingly important role cryptocurrency plays in the financial market but that there is relatively little existing empirical literature on cryptocurrency. This study provides reliable quantitative cryptocurrency market investigations to help fill this gap.
Citation
Shen, Ze (2021). Price Analysis and Risk Management in Cryptocurrency Market. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /195118.