Growth of Initial Investment Illustration
The first thing you will see at the top of the model preview page is an illustration of the growth of an initial investment in your model vs. the benchmark. You can use the "Edit Performance Chart Parameters" link to add additional benchmark to the graph, change the starting investment value, or set a new start date for the illustration. Additionally, you can hover over any point on the graph to see what the value of the model and benchmark(s) were on that date.
Hold the Ctrl key to select multiple benchmarks from the Popular Benchmarks box. You can also enter multiple tickers separated by commas in the Additional Benchmarks field.
If the colors of the lines on the chart are hard to see or differentiate, refresh the page and the graph will generate new colors.
Annualized Returns & Portfolio Statistics
The Preview screen will show the total cumulative return over the backtest period as well as annualized trailing returns for both your model and the benchmark. Additionally, the portfolio statistics shown will help you get a sense of the risk-adjusted returns of your model and its average annual turnover among other measures.
The performance you see here will not be exactly the same as the performance you see when you save your model. This is because in order to generate your model preview quickly, we run performance based on end of month prices over the backtest period. Once the model is saved, performance is calculated using daily pricing.
In certain cases you can tell if the model needs to be edited before saving it (for example, if the performance is far too low or if the turnover or standard deviation are far too high), but in general the best practice is to save your model so you can evaluate it with the most complete information. Additionally, once your model is saved you'll be given even more information to use in your evaluation (upside/downside capture, alpha, beta, etc.).
Model Holdings Over Time
This section shows every trade that occurred over the backtest period. Securities that appear in green with a + next to their ticker were added to the portfolio on that evaluation date. Securities that appear in red with a – next to their ticker were removed from the portfolio on that evaluation date. On the far left you can see the 10 stocks that were in the portfolio at the start of the backtest. Two months later XRX was sold and BHI was bought. The next month, AMZN and F were sold and COP and WY were bought.
This section can also provide you with an indication of whether your rules are too restrictive. This is particularly important if overlay filters are included. The below model is intended to hold 5 ETFs at all times. While the model does hold 5 funds the majority of the time, you can see that at times the model holds as few as 2 ETFs. If this is caused by an overly restrictive overlay filter, you can correct this by either altering your buy territory in the Strategy step of the model creation process to let more securities into your buy territory, or you can relax the filter criteria so more securities are likely to meet the overlay rules.
Lastly, this section can provide insight into how your Sweep Down is working. The below model is set to hold 5 funds, but it also includes a Sweep Down that will sweep assets into MNYMKT as it rises up the universe ranking. Below you can see that on 9/30/2022, 4 holdings were sold MNYMKT was added as a holding in their place. This means that at this point MNYMKT was 80% of the portfolio due to the Sweep Down function triggering. Subsequently, on 12/30/2022, MNYMKT was sold and 4 new holdings were added in its place, returning the model to being fully invested.
Model Allocation Over Time
This section is hidden by default, but clicking one of the links exposes graphs illustrating how your model was allocated throughout the backtest, broken down by asset class, investment category, or sector. This can provide a helpful illustration of the dynamic nature of your model, especially when juxtaposed with a more passive strategy. It can also help you determine if allocation maximums might be necessary if the model appears to become overconcentrated in certain areas of the market over the backtest period.
These allocation charts will rely on how a security is classified on its own and will not drill down further into a fund's holdings to determine its underlying allocation. For example, QQQ will appear as US Equity in the Asset Class Exposure chart and US Large Cap in the Investment Category Exposure chart but it will appear as N/A in the Sector Exposure chart since QQQ does not have a sector classification of its own.
Annual Returns
The last section of the preview page is annual returns relative to the benchmark. It may be helpful to study this to see if your model seems to perform as you would expect in certain years based on your rule set or intentions, or if you can glean anything from what might have driven performance based on the specific market environment of that year.
Saving Your Model
Once you are satisfied with your model, click the green Save Model button in the top right of the screen. If you want to make alterations to the model prior to saving it, hit Exit Preview.
Once you click Save Model, you'll be presented with a window where you can review the model rules, finalize your model name and ticker, and add a description of the purpose of the model for future reference.