This dissertation studies intraday and daily foreign exchange market volatility. First, we address how best to model the intraday seasonality and the serial correlation in return volatility. We find there is no gain from smoothing the intraday seasonal volatility pattern. A model that jointly estimates the intraday seasonal pattern and conditional heteroskedasticity underperforms models that remove seasonal variance through deseasonalization and then model conditional heteroskedasticity with a GARCH model. Secondly, we show how intraday data can be used to create daily volatility estimates. Results show intraday data allow for daily volatility estimates which are independent of a volatility dynamics specification. Lastly, we show that intraday data improve the performance of one-step ahead forecasts based on a one year sample and show that the results are consistent with Monte Carlo simulations.
| Identifer | oai:union.ndltd.org:arizona.edu/oai:arizona.openrepository.com:10150/288791 |
| Date | January 1998 |
| Creators | Sugiyama, Alexandre Borges |
| Contributors | Oaxaca, Ronald L. |
| Publisher | The University of Arizona. |
| Source Sets | University of Arizona |
| Language | en_US |
| Detected Language | English |
| Type | text, Dissertation-Reproduction (electronic) |
| Rights | Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author. |
Page generated in 0.3515 seconds