Work detail

Modeling the Czech Stock Market with High Frequency Time-Series Methods

Author: Mgr. Vít Bubák
Year: 2005 - summer
Leaders:
Consultants:
Work type: Finance, Financial Markets and Banking
Masters
Language: English
Pages: 102
Awards and prizes: The Czech Head Award
M.A. with distinction from the Dean of the Faculty of Social Sciences for an excellent state-final examination performance and for an extraordinarily good masters diploma thesis
Link:
Abstract: The first part of the thesis provides a detailed description of the microstructure of Prague Stock Exchange (PSE). Understanding the mechanism of trading on PSE serves as a necessary starting point for the main subject of the thesis –the empirical modeling of high-frequency data under the predictive framework of information-based models of market microstructure. As part of such understanding, we also analyze basic properties of high-frequency data using a sample of securities traded on the exchange. The next part of the thesis focuses on price duration (PD) process. Using a set of three of the most liquid securities traded on PSE's main market (SPAD), we first examine whether the intensity of bid-ask quote arrivals carries any information about the state of the market. The preliminary empirical analysis provides evidence of clustering effect in PDs: i.e., short (long) durations tend to be followed by short (long) durations, respectivelly. In fact, we show that larger autocorrelations in PDs tend to persist even after the time-of-day effects (seasonality) have been removed from the process. We take the analysis a step further when we test the predictions of market microstructure models using several proxies (intensity of trading, average volume per trade, and average spread). In abstract, our results tend to favor the conclusions of information based models, although any straightforward judgements remain at best ambiguous. In the last section, we turn our attention to the information content of a trade and try to measure the ultimate impact on a stock price of the component of the trade which is unexpected. In short, we find that (a) the full impact of a trade on the security price is not felt instantaneously but with a protracted lag; (b) as a function of trade innovation size, the ultimate impact of the innovation on the quote is nonlinear, positive, and increasing, but concave; and (c) the order flow is affected by prior quote revisions.
Downloadable: Vít Bubák
August 2020
MonTueWedThuFriSatSun
     12
34

56789
10111213141516
17181920212223
24252627282930
31      

Partners

Deloitte

Sponsors

CRIF
McKinsey
Patria Finance