Table 1: Total Investment of Pension Funds and All Retirement Vehicles, 2015 (Continued) 2012-2016; 2013-2016). In this way, we have included as much as possible pension funds to the study. Another aim of the study is to evaluate the performance of each individual fund companies over different window sizes, in order to see the changes in ranking positions them. The number of funds for each company and descriptive statistics for different research windows are shown in Table 2s Table 2. Descriptive Statistics of Different Windows (Continued) It is widely accepted that performance evaluation should consist of two components: risk and return. In constructing a measure of performance, determination of risk is the first important issue; either the total risk or the systematic risk. The second issue is how to combine risk and returm to construct the portfolio performance measure (Moy, 2002: 226). There are three general classes of performance measures dependent on the utilization of risk. The first performance measures based on the total (standard deviation) risk of return. The second class is comprised of systematic (beta or covariance) risk of return. The third class does not require a risk pricing model (Jobson and K orkie, 1981: 890). In this paper, we applied five different performance measurements, which based on both standard deviation and retum (Sharpe and Sortino Table 3. Performance Measurement Techniques
