Empirical analysis of trade duration distributions: The WIG20 case
Empirical analysis of trade duration distributions: The WIG20 case

Summary/Abstract: The aim of the study presented in this paper is to analyse the distributions of trade durations for WIG20 stocks using data from May 2025, with a particular focus on modelling doubly truncated data. Left-truncated distributions for trade durations have already been described in the literature, which is justified, as the values of an excessive proportion of observations were equal to zero. In this study, it is assumed that the data are also righttruncated due to time limitations between the trading sessions. Three doubly truncated continuous distributions were analysed in the study, namely the lognormal, the Pareto and the Weibull distribution. To satisfy the assumptions of stationarity and independence, the data were divided into smaller subsamples. Goodness-of-fit tests were then performed to determine which theoretical distribution best describes the empirical data. The results indicate that the quality of the fit depends on the lower truncation level – the higher the truncation threshold, the better the lognormal distribution fits the empirical trade durations.

• Details
• Contents