Revisiting the Stealth Trading Hypothesis

The stealth trading hypothesis (STH) states that informed traders concentrate their trades on medium sizes to conceal their information. Supporting empirical evidence for the NYSE case is considerable. In this paper, we question several methodological aspects of previous studies to provide new insights. We show that the STH cannot be rejected under alternative definitions of the trade-size cutoffs, after we control for bid-ask bounce, time between trades, and prevailing spread and depth. When we extract the friction-related component in price changes, we cannot reject the STH for some subsets of medium-sized trades, but we show that most of the disproportionally large role in the cumulative price change previously attributed to medium-sized trades dissipates.