New models for irregularly spaced time series analysis with applications to high frequency financial data

In financial econometrics data is being analyzed at higher and higher frequencies. Engle (1996) discussed ultra-high frequency data when all transactions are recorded. These may be transactions which occur in financial markets all around the world. The main characteristic of such high frequency data is that it is fundamentally irregularly spaced. Standard econometric techniques require the aggregation of such data to some fixed time interval. Clearly, if a short time interval is chosen, there will be many intervals with no information, on the other hand if a long interval is chosen the microstructure features of the data will be lost. Hence, it seems useful to develop methods which consider irregularly spaced data. In the empirical application I use transaction data that was abstracted from the TORQ data set for Dresser Industries share. The data set contains information about each transaction occurring on the consolidated market during regular trading hours over a three month period