The details of requirement: The final task would be: need to reproduce the R-code of the paper: A Test of Covariance-Matrix Forecasting Methods. Which means that use 5 method that mention on the paper: [login to view URL] Historical Covariance. [login to view URL] Historical Covariance With Shrinkage. [login to view URL] Weighted Moving Average. 4. Dynamic Conditional Correlation GARCH 5. Generalized Orthogonal GARCH. to forecast the covariance-matrix and then evaluate the performance of these covariance-matrix by their ability in out-of-sample tests to minimize the variance of portfolio. (We do not need code the last difficult part: The Global Minimum-Variance Portfolio with a Volatility Target) Require experience in time series analysis model and financial data!!