Here you find practical information on how to use the Adaquant Asset Allocation Tool. If for any reason you don’t find answer to your question here, use the contact form in https://www.adaquant.com/contact/ to submit your question to the Adaquant team. We will reply to you as soon as possible, and in case the question benefits other users, we add it also to this Frequently Asked Questions page.
First download the software installation file. You can find it from the website page “Software”. Scroll down to Free Trial section and hit the “Download” button. The file is then downloaded to your computer. The name of the installation file is Adaquant-AAS-Setup-x.y.exe, where x.y is the version of the software.
Note that the installation file for the free trial and any licensed version is the same. So you do not need to reinstall the software in case you buy a license and you have already installed the free trial version. It is therefore always a good idea to install the free trial version first before buying the license. This makes sure that the software works well in your computer and you get a good picture of the functionality of the software before buying the license.
Now go to the folder where you downloaded the installation file. Execute the file.
If you are executing the file in Windows 8 or 10 environment and you use Windows Defender, you may get the following notice: “Windows Defender SmartScreen prevented an unrecognised app from starting. Running this app might put your PC at risk.”. Don’t get discouraged by this! The reason is that the installation package is signed with a standard personal code signing certificate. Unfortunately the other types of certificates are only available to companies. This problem will automatically disappear once there are enough downloads and executions of the installation file. Until this, you can safely select More Info / Run Anyway option.
The first window that opens is the license agreement. It contains the licensing information for using the software. Read this through carefully and then accept the agreement.
Next window is general information. You may use the free trial for 7 days. To be able to use the software, you need to have Microsoft .NET framework 4.6 or newer in your computer. If this framework is not yet installed in your computer, there are two options: 1) if you are running the installation file with Administrator rights, the framework is automatically installed during this installation process or 2) if you are not running the installation file with Administrator rights, then you need to separately get the .NET framework installed to your computer and after that run the Adaquant Asset Allocation installation file again.
In the next window, the installation path is asked. If you are running the installation file with Administrator rights, the “PROGRAM_FILES_x86_DIR\Adaquant Asset Allocation Suite” folder is recommended as installation path. If you don’t have Administrator rights, then the recommended folder is “USER_HOME_DIR\AppData\Roaming\Adaquant Asset Allocation Suite”. Nevertheless, you can change the installation folder to be whatever you want.
Next window asks if you want to create a desktop icon for Adaquant Asset Allocation Suite.
After this you are ready to launch Adaquant Asset Allocation Suite. Have fun!
Buying a license to use Adaquant Asset Allocation Suite allows you to use version 1 of the software for unlimited period. Also you will receive the current quarterly Adaquant estimates for asset class returns, volatilities and costs for an ETF investor. If you have already read the book “Investing for Non-Finance People”, you know that it is especially now important to have valid estimates to make your asset allocation plan. This is because the expected returns of the asset classes are very different from the historic ones.
You can buy the license from the Adaquant website page “Software”. Press button “Buy License” in the upper part of the page and it brings you to the section where you can buy the license. After the purchase, you will be opened Download window, where you see your Software License Key (format XXXX-XXXX-XXXX-XXXX) and you can also download the latest Adaquant estimates file. You can also download the Adaquant Asset Allocation Suite installation file from this Download section, however, if you have already installed the free trial, there is no need to reinstall the software. You will also receive an e-mail containing your Software License Key and link to the Download section. If in some strange case your Software License Key is empty, notify us by using the Contact Form in the website. We will regenerate the license and send it to you.
Now, if your free trial period has already expired, when you open the Adaquant Asset Allocation Suite, Licensing window is automatically opened for you. If your free trial is still running, you can open the Licensing window from the menu “Help / Licensing”. In this window, you need to enter your e-mail address and Software License Key in order to activate the license. The e-mail address needs to be exactly the same as you used in the purchase process. Otherwise the activation will not work.
After inputting your e-mail and Software License Key, hit “Update License” button. If everything is correct, you will receive confirmation “License check successful. You can now use the full functionality of the software”. If not, and you have done everything correctly, notify us through the Contact Form in the website and we will help you through the license activation process.
Next step is to install the latest Adaquant estimates file. There is own question for this in the Frequently Asked Questions page: “How to install Adaquant estimates file?”
The next step after activating the license is to install the Adaquant estimates file. After the license purchasing process you got a window where to download the estimates file. You also received a link to this download area by e-mail. The estimates file name is estimates_YYYYQX.ada, where YYYY indicates the year of the latest estimates in the file and QX the quarter (Q2 means for example second quarter).
Actually the estimates file includes much more than just the latest estimates from the Adaquant team. It also contains cost profiles for ETF investors, updated asset class returns and updated ETF product lists. With all this you data you can create very sophisticated asset allocation plan to yourself and easily get an idea in which exact products you can invest.
So you have downloaded the .ada estimates file to some folder in your computer. Open Adaquant Asset Allocation Tool and select from the menu Help / Install Estimates. Small Window “Install Estimates File” opens. Click button “Select Estimates File”. In the file dialog you can now find the folder where you downloaded the .ada file and open the file.
Progress window opens and shows the status of installing the estimates file. When everything goes smoothly, you get confirmation “New data updates where successfully installed”. If something goes wrong, which should be very rare case, you get error notification. In this case contact us through the website Contact Form and we help you through the installation process.
You now have the latest data available to make your asset allocation plans! To use the estimates in the simulation, in the main window with simulation settings, select as “Estimates Profile” one of the estimates profile (USD / EUR / CHF / GBP Investor) and the latest quarter.
Whatever asset allocation software you use, one of the key elements is to understand how the software produces the asset allocation plans. There are as many approaches as there are products, and there is no single correct way to do the asset allocation plans. Nevertheless, I think some approaches are better than the others. If there is no information given on the underlying technology of the software, it is impossible to judge if the approach is good or not.
The Adaquant Asset Allocation Suite creates asset allocation plans by creating a set of simulations. The simulations are made based on monthly asset class returns. The asset classes that are pre-configured to the Suite have returns that are taken from US indices that represent the asset classes. For example, the returns of the “Large Cap Equities” are the total returns of the S&P 500 index. All asset classes have returns since December 1997.
To describe the process in detail, let’s make an asset allocation plan with 10-years investment horizon. The steps needed to arrive into the asset allocation plan are the following:
- Historic asset class returns are first adjusted with the estimated return and volatility of the asset class. These estimates are given as input by the user and can be for example the estimates from the Adaquant team.
- Make a sample of 10-year asset class returns based on the adjusted asset class returns. This sample is created by picking randomly 120 monthly returns from the adjusted asset class returns between December 1997 and June 2017. Repeat this procedure to create x samples. You can think of these samples as scenarios of what kind of different returns the asset classes may have in the future.
- For each sample of 10-year asset class returns, calculate efficient frontier. Efficient frontier is the set of optimal portfolios that offers the highest expected return for a defined level of risk. You can think of it as set of investment portfolios that have different levels of risk from very low risk to very high risk. Each portfolio has certain weight for each asset class. For example, if an investment portfolio has high risk, it may have high weight for equities which is high risk asset class.
- After all the efficient frontiers are calculated, calculate the average weight of each asset class in all y investment portfolios. So in the end you have y asset class weights that range from very low risk to very high risk.
- Next step is to calculate what is the return and risk if the investor has exactly average risk tolerance, i.e. take the middle from the y asset class weights. By using these weights, calculate the return and risk (maximum drawdown) for each x samples. Then look if the results satisfy the targets (return and/or maximum drawdown) of the asset allocation plan. Let’s say target is that maximum drawdown must be better than -20%. Check the maximum drawdown of each x samples, and if there is one or more that have worse maximum drawdown than -20%, the tested asset class weights have too much risk. So select less risky asset class weights from the y asset class weights and repeat the test. If the first tested asset class weights would have satisfied the targets, then more risky asset class weights can be used. This iteration is then continued until the best returning asset class weights are found that satisfy all the asset allocation plan targets. This is the final asset allocation plan.
- The asset class weights of the final asset allocation plan are then applied to each of the x sample 10-year asset class returns. This gives the distribution for the expected return and risk for the 10-year investment horizon.
The benefit of this approach is that it does not rely on historic asset class returns as such, but by using the sampling builds different scenarios that can happen in the future. The only assumption that the procedure makes is that the monthly asset class returns move more or less together the same way as they have in the last 20 years i.e. that there is similar type of correlation between the monthly returns. However even here there is no assumption that the correlation is a static number but it can also vary greatly between different 10-year sample asset class returns (as it has varied also in real life for shorter periods). The procedure results into very diversified, common sense asset allocation plans.
The additional simulation settings are the following:
- “Block Period Months” means that when the sampling is made to create hypothetical time series in the simulation, instead of picking 1 month returns randomly, the picking can also be done so that many consecutive months are picked at one time. Asset class returns may sometimes have momentum, for example equities may have longer periods (bull and bear markets) where the prices tend to move to one direction at one time. Therefore I like to keep this “Block Period Months” as 12 months.
- “Sample Size” indicates how many hypothetical time series are created in the sampling process. If the sample size is small (below 100), the simulation execution is very fast, but if you repeat simulations with the same settings, the results will differ. If the sample size is very large, the simulation results are repeatable but they take a long time to run. Through trial and testing, I like to keep this setting in 1000, which is good compromise to run relatively fast simulations with more or less repeatable results.
- “Efficient Frontier Sample Size” indicates how many portfolios with different risk levels are calculated for each hypothetical time series. Again, having high number increases simulation time but small number causes randomness to the results. I find 100 is a good setting.
- “Include Human Capital”: the tool also supports human capital asset class. It is an asset class which value depends on the amount of your future savings (income minus spending). If you want to make asset allocation plans with human capital, you can enable that functionality here.
When you run simulation, it may happen that no feasible solutions are found at all.
This results usually if you have put the settings so that they contradict each other. For a simple example, you might have Main Asset Class setting so that the weight is between 50-70%, but the individual Asset Class settings inside the main asset class are so that the maximum allowed weight is below 50%. In this case the optimizer is not able to find a solution. But this is just a very simple example. There are many settings so it is not always straightforward to find out what setting contradicts each other. You should play around with the settings to find combination that works.
Another scenario where no feasible solutions are found is that two asset classes correlate very heavily with each other (correlation close to 1). In this case the optimizer gets confused. The right way would be to put your investments only under one asset class and not defining an asset class that heavily correlates with the existing asset classes. Anyways, the definition of asset classes is that they group together similar investments and don’t correlate with other asset classes.
Yet another case is that you are running very short period simulations. If the amount of months (setting Period Years * 12) is less than the number of asset classes you have included to the asset allocation plan, then the optimizer cannot find a solution for you. For example, you might be doing plan for 1 year investment horizon but include over 12 asset classes to the plan.
One feature of Adaquant Asset Allocation Suite is that it has a list of pre-selected ETFs which in our opinion are excellent choices for individual investors. They are low-cost, diversified and well represent the asset classes that the software uses. So how to find the ETF you are looking for?
Go to menu “Tools / Investments / Print”. It looks like this (click image to see it larger):
You have four pre-configured templates for investors in USD, EUR, GBP and CHF. Basically this means that we have put a tag to each product when it is a recommended product.
If you want to filter the results more, you can do this in “Select Product Attributes”. For example, let’s say you are USD investor and want to find a small cap equity ETF. Browse to the USD investor template. For “Category” select “Asset Class”, for “Values” select “Small Cap Equities” and for “Criteria” select “Include”. Then press “Add” button. You see that this additional criteria is added to the filter list and the results in the below table updated accordingly. You have now four products to select from. You can sort the products by clicking the headers in the result table.
If you now press button “Print”, you will get a pdf that includes these four products in the order that they are in the result table (click image to see the pdf):
In addition to have inclusive criteria to filter the results, you can also have criteria to exclude products in the result list.
If you go to menu “Tools / Investments / Manage”, you can also add your own products/investments. You can also create your own investment attributes and attach them to new and existing investments.
When you create new investments, you can use them in the portfolio tool (Tools / Portfolio) to add the investments into your portfolio.
Creating the Adaquant estimates is a puzzle that starts with making estimate for one asset class and then moving step-by-step to make estimates for all the asset classes. Here is description of the steps needed to create all estimates:
- First estimate the returns for USD based investors. Estimate time horizon is 10 years.
- Expected inflation can be calculated from the difference of the yields offered by inflation protected fixed-income securities and normal fixed-income securities.
- Expected returns for large cap equities is estimated with the expected inflation, historic real return of S&P 500 and valuation model based on P/E ratio for the S&P 500 index
- The expected returns for small cap and emerging market equities is the expected return for large cap equities plus risk premium – the expectation is that small cap and emerging market equities will have slightly better return in long-term than large cap equities. Similar way additional return premium is used to estimate returns for private equity asset classes, because they have been outperforming S&P 500 in the long-term.
- For fixed-income asset classes, the yield-to-maturity of the asset class indices or ETFs forms the base for the expected returns. There is an estimate for default rates and the yield-to-maturity values are adjusted accordingly. This is especially important with high-yield debt where the yield-to-maturity number assumes there will be no defaults with the portfolio holdings. There are no assumptions where the interest rates will move in the future.
- The starting point of expected return for real estate is the expected inflation. However this is only the expected price appreciation of directly owning real estate. On top of this there are usually additional returns from renting the property and using leverage in the investment. So to calculate the final expected return, property specific calculation needs to be done. This calculation needs to be done separately for each investor because it depends on the properties which the investor owns or is targeting to buy. See How to model real estate investment returns for more information.
- For commodities including gold expected inflation is the estimated return.
- For bank accounts different bank account interest rate comparison websites are used to make the estimate.
- Now that there is an estimate for USD based investors, the estimates are done for other currencies. For some asset classes, the expected returns are derived from the USD estimates by using information of the interest level differences between the currencies. For real estate the expected return is the inflation specific to the currency and for the fixed-income investments the yield-to-maturity of the asset class indices or ETFs of that currency.
- Historic volatility is used as estimate for the expected volatility of the asset classes.