Page 299 - Contributed Paper Session (CPS) - Volume 2
P. 299

CPS1863 La Gubu et al.
                Based on portfolio weights, mean vectors and covariance matrices, we can
            then determine the return, risk and Sharpe ratio of the two portfolio models
            as presented in Table 5.

                Table 5 Return, Risk and Sharpe Ratio of Classic MV Portfolio and Robust FMCD
                                            Estimasi Portfolio
                          Return                  Risk                Sharpe Ratio
              
                                              
              0.5   0.0430095  0.1022346   0.0856715   0.2060775  0.1464506  0.2248905

               1   0.0216108  0.0507269   0.0214754   0.0515543   0.1464871  0.2227779

               2   0.0109114  0.0161286   0.0054264   0.0085656   0.1461718  0.1727134

               5   0.0044918  0.0070372   0.0009326   0.0016471   0.1423737  0.1698539

              10   0.0023519  0.0032753   0.0002907   0.0004736   0.1295127  0.1438972

            4.  Discussion and Conclusion

            4.1 Discussion
                From the results obtained through cluster analysis, it was found that LQ-
            45 stocks can be grouped into 7 clusters as shown in Table 2, where each
            cluster had a  different number of stocks. In cluster 1, BBCA have the best
            performance compared to other stocks in the cluster which are marked with
            the highest Sharpe ratio in the cluster, which is 0.0523. So that BBCA are taken
            as a representation of cluster 1. Furthermore, in cluster 2, BRPT stock with
            Sharpe ratio 0.00953 are representations of cluster 2. And so on, WSKT stock
            with Sharpe ratio  -0.01033 are representations of cluster 7. Therefore, it is
            enough to consider stocks as presented in table 3 for investment decisions.
                From Table 4, it can be seen that stocks with negative returns namely AALI
            and WSKT have a negative weight (short selling) for all risk aversion values 
            both in the classic MV portfolio model and in the MV portfolio model with
            robust FMCD estimates. Conversely, stocks with large returns, namely BBCA
            and INCO stocks always have positive weights in both portfolio models.
                Measuring  portfolio  performance  not  only  base on  the  return  but  also
            must pay attention to the risks that will be borne by investors. There are several
            measures that can be used to measure portfolio performance, one of which is
            Sharpe ratio. Table 5 shows the return, risk and Sharpe ratio of the portfolio
            formed using the classic MV portfolio model and portfolio model with robust
            FMCD estimation. From Table 5, it can be seen that for  = 0.5, portfolio return
            with robust FMCD estimation is 0.1022346 and risk is 0.2060775, while classic
            MV  portfolio  return  is  only  0.0430095  and  risk  is  0.0856715.  So  the  MV


                                                               288 | I S I   W S C   2 0 1 9
   294   295   296   297   298   299   300   301   302   303   304