This paper empirically investigates the effects of a particular type of illiquidity on asset finding the average illiquidity discount is around 77%, and the transaction of restricted shares slightly decreases the price of otherwise identical freely-traded shares, by 0.61%. We build an empirical model to study how the discounts vary with firm-specific and transaction-specific variables finding that illiquidity discounts are positively related to the length of trading constraints, and volatility. In the end, we make a comparison of discounts from empirical and theoretical models (Longstaff 1995 and 2001) and propose some explanations for the difference between them.