| Purdue Computational Finance Program | |
|
Analysis of High-Frequency Data
February 25, 2003 |
|
|
Abstract:
High-frequency intraday financial data provide empirical researchers with an unprecedented view into the inner workings of financial markets. Such data sets often contain detailed information regarding the nature of individual transactions and quote updates, not unlike the full information set used by market participants. As a result, much can be learned about how the market learns about new information and how this new information becomes impounded into asset prices. With these new data sets, however, come challenges for the empirical investigator. Since transactions generally do not occur at fixed intervals this chapter studies the analysis of high frequency financial data from the view point of a marked point process. Models of marked point process that are particularly well suited for the analysis of high frequency data are presented. We also consider some implications of converting irregularly spaced data to fixed intervals for the purpose of performing traditional fixed interval analysis. |
©
2003 Purdue University Last Update: Feb 20, 2003 Please send comments and suggestions to the Webmaster. |