Optimizer 5.4 is a Windows based application which given a desired portfolio composition, using investor profiles and under specified constraints, will provide you with the best allocation, taking into account the nonlinear correlation among your stocks and minimizing the CVaR given a target return and risk profile.
Download the program for free from this link and follow the installation instructions. Once installed, start the program by double clicking on the “Optimizer 5.4” icon (if you chose to make a desktop icon, if you didn’t, you will find it among your programs, in program files)
Once the interface appeared, click on the “Portfolio Composition” button
A browser interface will show up giving you the choice among the 500 tickers from the S&P 500 Index. Hold Ctrl and select the tickers you would like to have in your portfolio then press “Open”. There is no minimum nor maximum size, but be aware that for a very large portfolio you need a powerful PC, so use the utility at your discretion. If the size will be very large the data processing and calculation will take some time. You are free to add your preferred yahoo ticker (if not present by default) in the program’s installation folder:
e. g. C:\Program Files (x86)\Optimizer
Note: the tickers should be a .mat file with the ticker text only
After having chosen your portfolio size, press now the next button which is the “Get Quotes” one.
A new dialog box will appear asking you to insert the date your analysis should be done from. The date format has to be ddmmyyyy.
Press OK and the adjusted closed stock prices for the selected tickers will automatically be downloaded from the yahoo finance server from the beginning date indicated and till today.
The next button is called “Analytics” and it plots the cumulative performance, the returns, the ACF and PACF of the Equally Weighted portfolio formed by these stocks.
If the period is long enough we can notice the AR(1) effect from the ACF function and sometimes MA(1) effects as well. Next we press the “Model” button and a dialog box will pop up asking you for your risk profile. In order to choose your risk profile, type 1 for Conservative risk profile, meaning that you aim at an expected return of 0.75 times the Equally Weighted portfolio return, but less risky; type 2 if you want to choose the Core risk profile and to be under the constraint of minimizing the risk for a targeted return equal to the Equally Weighted portfolio. Finally, type 3 if you are a risk seeker and choose the Growth risk profile aiming at an Expected Return equal to 1.25 times the expected return of the equally weighted portfolio.
Together with the model the bivariate scatter plots of the first two assets copula functions will appear in order to have a look and to compare the various parametric copulas with the empirical one. As you will notice, in every experiment, the t copula will be the closest one to the empirical copula. This is just a simple graphical representation; no test on goodness of fit of copulas is done. The t copula is chosen by default because of its success along multiple studies to simulate the financial returns nonlinear dependence and because of its easy implementation.
The next button and the last one is the “Optimization” button
Once pressed it will ask you about the CVaR alpha accuracy level you prefer. The supported values are from 0.001 to 0.1
Once you inserted the alpha level, press OK and the optimization process will start. Depending on your portfolio size it will take a few seconds or minutes and it will output the optimal weights for the given stocks. This will be a portfolio minimizing the CVaR and taking into account the nonlinear dependence among them, the ARMA-GARCH effects and also under the Extreme Value Theory.
At the same time, you will get the corrected VaR and CVaR of the optimal portfolio and the daily and monthly expected return.
Finally click on Export to generate an Excel table with the optimal portfolio weights’ output and then press “Open File” to preview. For a new portfolio selection repeat the same steps as above.
Please note that the results will depend on the period analyzed, on the investment profile of the portfolio holder and of the investor’s risk aversion. Kindly be advised that even if the utility takes into account the adjusted returns, it is not taking into account the transaction costs which are high in the case of frequent re-balancing portfolio.