Testing Stability of Correlations Between Liquidity Proxies Derived from Intraday Data on the Warsaw Stock Exchange

The aim of this paper is to investigate relationships based on correlations between alternative liquidity measures derived from intraday data on the Warsaw Stock Exchange (WSE). Analyses of correlations help to find an answer to important question whether different liquidity proxies capture various sources of market liquidity/illiquidity or not. The main research hypothesis states that correlations are stable in specified periods. The hypothesis is verified by applying equality tests of correlation matrices computed over non-overlapping subsamples. The dataset consists of daily proxies of four liquidity measures for 53 WSE-traded companies divided into three size groups. The whole sample covers the period from January 3, 2005 to June 30, 2015, and it includes three adjacent subsamples, each of equal size: the pre-crisis, crisis, and post-crisis periods. To calculate several liquidity measures based on intraday data it is essential to recognize a side initiating the transaction and to distinguish between buyer- and seller-initiated trades. In this paper, the Lee and Ready (J Financ 46(2):733-746, 1991) trade classification algorithm is employed to infer trade sides. The obtained empirical results concerning the stability of correlations between liquidity proxies in the whole group of companies are not homogeneous.