Abstract: | The investment performance of a Canadian portfolio of Canadian public real estate companies is analyzed over the period 1971–79. Using Sharpe's index of performance, it would appear at first sight that this portfolio exhibited remarkable superior performance. Moreover, this conclusion is not due to some peculiarity of the Sharpe measure: results using Treynor's measure suggest a similar conclusion. We then apply the significance tests recently recommended by Jobson and Korkie (1981). When their preferred test is applied, we are unable to reject the null hypothesis that the real estate portfolio did not exhibit superior investment performance. This result illustrates the necessity of performing adequate statistical significance tests whenever investment performance is being evaluated. |