Load Portfolio

Load a vector of expected returns and a covariance matrix for 30 stocks.

loadStockStats

CreatePortfolioObject with Default Constraints

The default constraints for aPortfolioobject are that it is a long-only portfolio and that it is 100% invested. Many other constraint types are possible.

P = Portfolio('mean',expRet,'covar',expCov); P = setDefaultConstraints(P);

Plot Full Efficient Frontier

Use theplotFrontierfunction with thePortfolioobject to visualize the full frontier. Other functions exist that allow you to probe into particular portfolios along the frontier.

P.plotFrontier

Capture Upper and Lower Bounds of Portfolio Risks and Returns

It is useful to know what are the upper and lower limits of the portfolio moments along the efficient frontier. This information allows you to determine what are feasible targets. Use theestimateFrontierLimitsfunction with thePortfolioobject to identify the weights at these extremes. Then you can use theestimatePortMomentsfunctionPortfolioobject to find the limiting moments.

P_Weights1 = P.estimateFrontierLimits; [P_risklimits, P_returnlimits] = P.estimatePortMoments(P_Weights1)

P_risklimits =2×10.0786 0.2868

P_returnlimits =2×10.0954 0.2370

估计有效Portfolio with Target Return

Select a target return for the portfolio somewhere in the feasible region. You can estimate its makeup withestimateFrontierByRiskand then confirm its moments usingestimatePortMoments.

targetReturn = 0.15; P_Weights2 = P.estimateFrontierByReturn(targetReturn); [P_risk2, P_return2] = P.estimatePortMoments(P_Weights2)

P_risk2 = 0.1068

P_return2 = 0.1500

The return matches thetargetReturnvalue and the risk is in agreement with the efficient frontier plot.

估计有效Portfolio with Target Risk

ThePortfolioobject is able to find an efficient portfolio with a specified target risk.

targetRisk = 0.2; P_Weights3 = P.estimateFrontierByRisk(targetRisk); [P_risk3, P_return3] = P.estimatePortMoments(P_Weights3)

P_risk3 = 0.2000

P_return3 = 0.2182

Another useful feature of theestimateFrontierByReturnandestimateFrontierByRiskfunctions is what happens if you specify an infeasible (too high or too low) target. In this case, these functions provide a warning message to suggest the best solution.