The notion of liquidity is widely used in the finance area, such as in the studies of market microstructure and asset pricing. However, there are so many liquidity measures that it is difficult for researchers to decide which measure they should adopt. There is no consensus on which measure is the most appropriate as well. This paper compares various liquidity measures to find which of these measures is the most appropriate in the stock market. Analyses are performed on the nonfinancial firms listed in the Korea Exchange and the NYSE/AMEX for the period 1993~2004. Two methodologies are employed. One is a correlation check that will reveal the internal consistency between measures. The other method is to investigate the relation between risk-adjusted returns
and liquidity measures by using the asset pricing framework of Brennan, Chordia, and Subrahmanyam (1998) (BCS). This paper provides us several kinds of new knowledge about liquidity measures. First, most liquidity measures are highly correlated with each other, except for the Pastor and Stambaugh (2003) measure. Therefore, we do not need to be severely concerned about the conflicts of the liquidity measures. Second, the Amihud (2002) measure and its modified measure perform distinguishably well. Third, researchers are interested in a liquidity measure which can replace the high-frequency measure, such as the bid-ask spread. This paper concludes that the most reliable solution is the Amihud (2002) measure.

