skip to primary navigationskip to content

Peter Malec Profile


Peter Malec's research focuses on the econometric analysis of financial high-frequency data. The importance of the latter has been growing over the past decades, which was triggered by an increasing intraday trading activity in modern financial markets, as well as advances in the technology for recording, storing and processing vast datasets. On the one hand, so-called intraday or high-frequency data offer great opportunities, which are mainly due to the substantial advantage in terms of the amount of information provided when compared to, e.g., daily observations of financial variables. As a matter of fact, the limiting case is a situation in which records on every single transaction or even every order event are available. On the other hand, researchers face numerous challenges, since high-frequency data exhibit features that are not encountered at lower frequencies. Hence, its analysis requires new tailor-made econometric methods. In this context, Peter's PhD thesis covers the dynamic and distributional modelling of high-frequency data, as well as its utilisation for risk reduction in vast-dimensional portfolios.

Peter's current research programme mainly deals with data from limit
order books, which consist of all standing buy and sell orders queued
according to price and time priority. For the econometric analysis, he mainly uses non- and semi-parametric techniques. Compared to classic parametric methods, the latter are considerably more robust as to the underlying set of assumptions. However, they require larger samples in order to ensure a comparable precision of the estimators. This fact makes the analysis of limit order book data a nearly perfect application as the latter allow for datasets of almost arbitrary size. More precisely, Peter employs the above class of methods to explain (discrete) movements of transaction prices and order book quotes based on a set of other order book variables, such as spreads or depths, as well as the trading activity of specific traders. Further, Peter aims
to develop more precise estimators of (low-frequency) asset return volatility employing information from limit order books.

« January 2017 »