# Portfolio Optimisation FRGS - Executive summary - (10%) 300 words - problem statement 2% - objectives 2% - methodology 2% - expected output 2% - significant of output 2% In this research project we will study the dynamic of cross-correlation of Malaysian stock markets. The analysis of cross-correlations is widely applied to understand the interconnections among stocks in a market and hence in the portfolio risk estimation. Current studies of correlations in Malaysian market mainly focused on the static correlations, and this calls for an urgent need to study their dynamic correlations. Our study focused to reveal the dynamic evolution of cross-correlations in the Malaysian stock market, and offer an exact interpretation for the evolution behavior. The objectives of the project are: (1) study and compute the time-varying cross-correlation of Malaysian stock data (2) build a mean-variance portfolio by using non-convex optimisation with regularisation (3) analyse the dynamics of the components of the optimal portfolio under different time regimes. The research activities that are expected to be carried out to achieve the objectives of the project are as follows: - formulate a method to compute efficiently time-varying cross-correlation of stock data (formula???) - infer time-varying covariance from cross-correlation and formulate a dynamic mean-variance portfolio - select suitable optimisation methods to minimize the portfolio. - Anlyse and describe the flucturations and trends for the components of the optimized portfolio. We are expected to gain insightful knowledge on changes of a optimal portfolio over the time. It is expected that the gained result would be useful for Malaysian traders to have better understanding on the changes of market risks and better risk projection. This will be invaluable to risk managers.