The accuracy of trade classification rules for the Warsaw Stock Exchange
Paweł Miłobędzki , Sabina Nowak
AbstractWe evaluate the accuracy of five trade classification rules for the Warsaw Stock Exchange: tick, reverse tick, quote, Lee and Ready and Ellis, Michaely and O’Hara. In doing so we use the transaction data on stocks from the large cap WIG20 index from the period May-September 2017. We find that the quote rule correctly classifies 100% of transactions initiated by buyers and sellers. Almost the same excellent job does the Lee and Ready rule. The Ellis, Michaely and O’Hara rule is less successful albeit its success rate exceeds 95% of the transactions assigned to both sides. The tick and the reverse tick rules exhibit a very low accuracy. The tick rule correctly classifies only 25.35% of transactions initiated by buyers and 25.95% of transactions initiated by sellers. The reverse tick rule performs even worse classifying as much as 16.66% and 16.67% of such transactions accord-ingly. The reason for their low accuracy is that the stock prices remain unchanged at the WSE at about 70% of all transactions. We also show that in case both classification rules are modified to account for either the preced-ing or the following transactions price changes their accuracy significantly increases.
|Publication size in sheets||0.59|
|Book||Papież Monika, Śmiech Sławomir (eds.): The 12th Professor Aleksander Zelias International Conference on Modelling and Forecasting of Socio-Economic Phenomena : conference proceedings, Socio-Economic Modelling and Forecasting, no. 1, 2018, Foundation of the Cracow University of Economics, ISBN 978-83-65907-20-2, 611 p., DOI:10.14659/SEMF.2018.01|
|Keywords in English||accuracy of trade classification rules, market microstructure, Warsaw Stock Exchange|
|License||Publisher website (books and chapters only); published final; ; with publication|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.