PORTFOLIO OPTIMISATION FOR MALAYSIA'S TOP 30 AND MID 70 ASSETS USING MEAN-VARIANCE MODEL

The goals of this research are to minimize the risk of losses for specified returns using the mean-variance model and to compare the risk and return valuations (in terms of in-sample and out-of-sample analysis) when the optimization is implemented on three different sets of assets. The assets consist of constituents of FBMKLCI, which represents the Top 30 Risky Asset and FBMM70, which represents the Mid 70 Risky Asset. The closing price data are drawn from Thomson Reuter Eikon. The mean-variance model is implemented using AMPL and the numerical results were analysed in Microsoft Excel. The general assumption on mean-variance is the higher the return, the higher the risk. Main findings show that the higher the expected return, the higher the risk at Top 30 Risky Asset. The number of assets that constructed the portfolios was more diversified as the risk decreased. While Mid 70 Risky Asset does not follow the general assumption of mean-variance. The combination of the two assets provides a more interesting outcome. The result improved in terms of the level of risk. The insertion of the really risky asset in a basket of assets somehow affects the behaviour of assets in terms of risk. We validate our in-sample portfolios by using out-of-sample analysis. The result shows that a combination of both Risky Assets gave better performance mainly for low and medium target returns.