On the correlation analysis of illiquid stocks

Abstract

The serial correlations of illiquid stock's price movements are studied, by taking into account for the effect of the stochastic zero returns in various situations. More precisely, heteroscedasticity and time-varying zero returns probability patterns are considered.Depending on the set up, we investigate how the usual autocorrelations can be accommodated, to deliver an accurate representation of the price changes serial correlations.An index and a cumulative sum (CUMSUM) test for choosing the adequate tool are provided. We shed some light on the properties of the different serial correlations measures, by mean of Monte Carlo experiments. The theoretical arguments are illustrated considering shares from the Chilean stock market.

Powers correlation analysis of non-stationary illiquid assets

“…In the same way, Raïssi (2020) found that the classical Box test tends to reject the no serial correlation for (r t ) erroneously when the zero returns probability is non-constant as in Assumption 4. For these reasons, we believe that the power correlation study can be carried out, using the (r t ) directly in most of the cases.…”