Type
Text
Type
Dissertation
Advisor
Kim, Aaron | Rachev, Svetlozar | Glimm, James | Xiao, Keli.
Date
2014-12-01
Keywords
Applied mathematics | generalized hyperbolic distribution, heavy tails, long memory, optimal order execution
Department
Department of Applied Mathematics and Statistics.
Language
en_US
Source
This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.
Identifier
http://hdl.handle.net/11401/77532
Publisher
The Graduate School, Stony Brook University: Stony Brook, NY.
Format
application/pdf
Abstract
Financial time series data exhibits heavy tailed, volatility clustering and long range dependence style facts. Traditional Gaussian distribution assumption based model failed to explain these phenomena. A unified framework model proposed in this thesis, fractionally integrated autoregressive moving average (FARIMA) and fractionally integrated generalized autoregressive conditional heteroskedasticity (FIGARCH) with multivariate generalized hyperbolic distribution (MGHD), trying to capture all these phenomena together. We also examined this model by using intra-day market dataset to backtest of various risk measure. With rise of high frequency trading and algorithm trading in recent years, trading volume hugely increased and markets became more volatile. Order execution is the main concern for traders, especially in the case of liquidation of big orders. We illustrate how the optimal order execution strategy behaves under the assumption that market price dynamics follows high volatile (non-Gaussian) markets with volatility clustering and log-range dependence characteristics. | 82 pages
Recommended Citation
Chai, Yikang, "Modeling Intra-day Markets with an application of Risk Management and Optimal Order Execution" (2014). Stony Brook Theses and Dissertations Collection, 2006-2020 (closed to submissions). 3333.
https://commons.library.stonybrook.edu/stony-brook-theses-and-dissertations-collection/3333