In this paper, an alternative method of estimating the systematic risk for Canadian stocks is presented and empirically investigated. The method proposed is applied to a set of data impacted by censoring - the presence of zero returns, which occurs in extreme cases of thin trading. The approach used is the sample selectivity model, which is a two-step procedure: with a selectivity component and a regression component. In addition, this study compares the new beta estimate to the standard OLS beta and the Dimson Beta. The results indicate that the selectivity-corrected beta does correct the downward bias of the OLS estimates and possesses desirable statistical properties.