Randomized Scenarios

What & Why

Run Randomized Scenario is one of two features included in the Stress Testing add-on. It allows you to generate a sequence of returns for your plan based on your expected average rates and historical volatility (or your preferred market assumptions). The other feature is Run Historical Scenario, which allows you to select a time period from the past and apply investment returns and inflation rates from that period to the current scenario.

We created Run Randomized Scenario to give you the ability to:

• generate and model sequences of returns with their ups and downs (instead of flat sequences, which are typically used for financial projections);
• show a range of outcomes anywhere between the worst and best case from a set of 101 randomly generated sequences;
• get an approximate indication of how likely your clients are to achieve their goals (you can use this feature to perform a basic form of stress testing);
• illustrate the importance of the order of returns to your clients.

If you’re eager to get started, you can jump straight to use cases, but you may want to learn the basics first. This article provides detailed information on the Run Randomized Scenario feature. If you prefer to see this feature in action, please watch our recorded webinar all about Stress Testing or see one of the use cases demonstrated in this quick 5-minute video. 

How it works

The basic workflow is to open the randomized modal, select a sequence of returns from a set of 101 randomly generated options, and apply investment returns and inflation rates from that sequence to the current scenario. Then you will be redirected to the Planning page, where you can see the rates applied to your scenario. You will also see some new assumptions, including a chart, on the Assumptions page in the report.

Randomized modal

You can enable this feature by selecting Run Randomized Scenario from the Planning page, which you can find under the Run Scenario dropdown. You may need to activate the feature.

This modal has everything you need to run randomized scenarios:

• A chart illustrating the growth of the three asset classes and inflation based on the randomized sequence of rates currently selected.
• Controls to help you select a different sequence of rates from the set of 101 randomly generated options.
• Buttons that allow you to apply the selection to the current scenario and rerun it.

Selecting a randomized sequence (i.e., set of rates)

You can select a specific set of rates using the following options.

The slider at the bottom of the chart allows you to pick from among 101 randomly generated sequences.

The initial 101 sequences are generated using your assumptions from the General page of Scenario Setup for Inflation, Cash, Fixed Income and Equity. These rates are used as the geometric average for each category and are combined with historical standard deviations and correlations to generate random rates of return and inflation for the length of your projection. More information about these calculations can be found in the additional notes at the bottom of this article.

The slider has a scale from 0 to 100. The left end of the slider represents the worst or lowest average real return while the right end of the slider represents the best or highest average real return. The initial positioning of the slider is 50, which shows the median average real return for the 101 sequences generated.

If you'd like to speak in terms of the probability attached to the range of outcomes going from the worst to the best case, if the slider is at position 50, you could say something like "Given our expected returns for investments and the historical volatility that we've seen in markets, you have a 50% chance of experiencing worse returns than we're seeing here and a 50% chance of experiencing better returns."

You can choose to proceed with the default sequence (i.e., 50), or you can change the slider to an alternate location on the bar. For instance, you could move the slider to the left until you reach 20 or 30. You could also move the slider until you find an average real return that you'd like to apply to the plan (e.g., 3%).

Once you select a different location on the slider, you'll need to select Update to update the graph and show fine print sections of the modal.

At position 20 on the slider, you could say "Given our expected returns for investments and the historical volatility that we've seen in markets, you have a 20% chance of experiencing worse returns than we're seeing here and an 80% chance of experiencing better returns."

If you would like to generate a new set of 101 rate sequences, you can click Randomize Again at the bottom of the modal. This may be useful in several cases:

• You would like another set of rate combinations to review because you're looking for a particular sequence (e.g., negative investment returns at the start of the client's retirement, multiple large market corrections within the sequence)
• You would like to illustrate to your client that the average case can result in different outcomes for their financial future depending on the sequence of future returns.
• You have updated the show fine print expected rates or standard deviations and would like a new set of sequences based on these updates.

Worst, average, and best buttons

The three buttons in the bottom-left corner of the modal called Worst Period, Average Period, and Best Period are there for convenience. All that they do is move the slider to the lowest (0), middle (50), and highest (100) positions, respectively.

Back to top

Show fine print

You can select Show fine print at the top of the modal to learn more about the feature, including assumptions and calculations. This section also shows the Actual geometric average annual rates that are reflected in the graph for each individual category of Inflation, Cash, Fixed Income, and Equity.

The correlations show the relationship between each individual rate and whether they move together (positive), apart (negative) or randomly (zero) with one another throughout the sequence.

Customizing expected values

From this section, you can also customize the average rates and standard deviations that are used to generate the random sequences. This is not a necessary step and is a more advanced option for users who would like greater control over the inputs. To make changes, you can select show Expected in the top-left cell of the Actual table.

You will then have the ability to change the Average Rate or Standard Deviation for each category.

Back to top

Align actual with expected

It was mentioned above that each of the 101 sequences generated with the standard Randomized Scenario feature is randomly generated. Therefore, it is unlikely that the Actual averages and standard deviations for each category align with the Expected values.

However, there may be cases where you want to illustrate a plan for your client with specific Actual results while maintaining the variability of a randomly generated sequence. In this case, you can select Align actual with expected in the bottom right corner of the modal and then click Randomize Again.

This functionality works by generating a random sequence similar to the 101 sequences we discussed above. Then Snap applies adjustments (e.g., increase/decrease all values to adjust the average rate, stretch/compress values to adjust the standard deviations) to the individual rates to create a sequence with the desired characteristics. There may be cases where the Actual values are different than the Expected ones desired. In those cases, you can click Randomize Again to generate a new sequence or you can reach out to our Customer Success team at [email protected] to review the case in more detail and discuss options to meet your desired outcome.

To generate a new set of 101 random sequences with the slider enabled you can deselect Align actual with expected and then click Randomize Again.

Invert expected averages

If you have the Align actual with expected box selected, you can also select Invert expected averages.

You will use the green navigator section below the chart to select your desired period.

You will see the period that you selected included below the Invert expected averages line.

Please ensure that your selected period is a relatively small length of the projection (e.g., 2 to 5 years out of a 40-year plan). Since Snap is trying to achieve both inverted rates for the selected period as well as your overall target rates for the full projection length, if your selected inverted period is too long you may not see the desired impact.

If you have any questions about these more advanced options, please feel free to reach out to our Customer Success team at [email protected].

Back to top

Running Randomized Scenarios

When you're ready to apply your selection and run the calculations, click Apply & Run in the bottom-right corner of the modal. This will apply your selection to your plan.

Overrides

If you run a Randomized Scenario on your plan, you won't be able to edit rates of return on the Assets page and you will not be able to override rates of return on the Planning page anymore. This is because randomized rates will be applied instead.

However, you can still edit and override allocations for any Financial Asset in any year of the projection.

Incomes indexed with General Inflation

After selecting Apply & Run, any incomes that have Indexing set to General Inflation will vary with the annual randomized inflation rates. (To specify General Inflation under the Indexing column, delete the indexing percentage in the cell. This will populate the field with General Inflation.)

Clearing randomized data

If you want to remove all randomized data and go back to a normal scenario, you have two options:

• Go to the randomized modal and click Clear & Run in the bottom-right corner of the modal.
• Go to the Assets page and click Clear Randomized Returns.

Look for the orange buttons in the screenshots above. They will restore your original Financial Asset return assumptions (except for overrides on the Planning page) and recalculate the scenario.

Back to top

Use cases

Please be sure to read these important notes first. Here are some common examples of how this feature can be used. If you find a different way to use this feature, we would love to hear from you!

Natural sequences of returns

If you want to find and model a randomized sequence of investment returns and inflation rates, you have a couple of options:

A. Select a sequence from a range of 101 randomly generated rates:

1. Open the randomized modal.
2. Update the asset mix to your client's approximate weighted portfolio asset allocation (e.g., 30% fixed income and 70% equity) and select Update.
3. Adjust the slider until you find a location (e.g., 20, 30, 50) or an average real return (e.g., 2%, 3%) that you would like to model.
4. Apply and Run the scenario.
5. Review the Planning page and report.

B. Select a randomized sequence that has actual average returns that align with your expected returns for each category (i.e., Inflation, Cash, Fixed Income, Equity).

1. Open the randomized modal.
2. Enable Align actual with expected.
3. Click Randomize Again.
4. Apply and Run the scenario.

Market declines during a specific period (e.g., beginning of retirement)

If you want to find and model a randomized sequence of investment returns and inflation rates that illustrate a market decline during a specific period, you have a couple of options:

A. Select a randomized sequence that has actual average returns that align with your expected returns for each category (i.e., Inflation, Cash, Fixed Income, Equity) and a specific period that has inverted expected rates of return.

1. Use the Planning page to determine what calendar year your client is shown to retire.
2. Open the randomized modal.
3. Enable Align actual with expected and Invert expected averages.
4. Drag the navigator handles (bounding the green selected period) so that your client's retirement period from step 1 is highlighted (e.g., starting at the beginning of the year of retirement through to 2 years after).
5. Click Randomize Again.
6. Apply and Run the scenario.

B. Select a sequence from a range of 101 randomly generated rates:

1. Open the randomized modal.
2. Update the asset mix to your client's approximate weighted portfolio asset allocation (e.g., 30% fixed income and 70% equity) and select Update.
3. Adjust the slider for different sequences within a range that you're comfortable with (e.g., between 40 and 60) and Update after each change to the slider location.
4. Review the chart to see if investment returns around your client's retirement period appear negative.
5. Apply and Run the scenario.
6. Review the Planning page to confirm the investment returns applied during the first few years of retirement.
7. Go back to step 3 if necessary.
8. If you're unable to find a sequence within your desired range on the slider (e.g., 40 to 60) you can Randomize Again to generate 101 new sequences to pick from.

Basic stress testing

If you want to find out how likely your clients are to achieve their goals, you can use this feature to perform a basic form of stress testing by running multiple randomized scenarios and making simple manual adjustments between the runs.

• Open the randomized modal.
• Update the asset mix to your client's approximate weighted portfolio asset allocation (e.g., 30% fixed income and 70% equity) and select Update.
• Adjust the slider until you find a location (e.g., 20, 30, 50) or an average real return (e.g., 2%, 3%) that you would like to model.
• Apply and Run the scenario.
• Review the Planning page to see if the client runs out of money before the end of the plan.
• Open the randomized modal again.
• Select Randomize again to generate a different sequence with a similar average real return.
• Apply and Run the scenario.
• Review the Planning page to see if the client runs out of money before the end of the plan.
• Repeat this process, randomizing the scenario each time to see in how many cases your client's plan is successful.

There are many ways to quantify the likelihood of success in stress testing. This is only one of them. Another option is available through our Historical Scenario feature that's included in the Stress Testing add-on. Which solution to use will be based on your and your client's preferences. Using a historical context is often easier to understand and explain to clients. Using a random sequence provides more flexibility to generate a particular sequence with your target rates of return and volatility.

In extreme cases, none of the available returns will be able to make the scenario successful or unsuccessful.

• If the slider is set to the lowest position (i.e., 0) and the scenario is successful, it means even with the lowest average real return generated in 101 random sequences using the input averages, standard deviations, and correlations your client would make it.
• If the slider is set to the highest position (i.e., 100) and the scenario is unsuccessful, it means even with the highest average real return generated in 101 random sequences using the input averages, standard deviations, and correlations your client would not make it.

Please remember that in real life countless factors can affect your clients’ ability to achieve their goals. Investment returns are only one of them!

Order of returns

If you want to illustrate the importance of the order of returns to your clients:

• Open the randomized modal.
• Enable Align actual with expected.
• Apply and Run the scenario.
• Rename the scenario to something like "Randomized Scenario 1”.
• Copy the scenario and rename the copy to something like "Randomized Scenario 2".
• Open the randomized modal for the copied scenario.
• Select Randomize Again and then Apply and Run the scenario.
• Compare the outcomes of the two scenarios.

This will display that even if the client experiences the same average annual rates for each category (i.e., Inflation, Cash, Fixed Income, and Equity) their outcome can vary significantly depending on the order of these rates.

Back to top

Reports

When you open the report, and look at the Assumptions page, you will notice some differences between regular and Randomized Scenarios.

For regular scenarios, we show the default rates of return per asset class (these are the rates that are used as the default for new assets). For randomized scenarios, we show average rates of return instead. See the right column. These averages are calculated from the resulting scenario that's created after you select Apply & Run. They may not be the same as the average returns you see in the randomized modal – those averages are for the years in the historical period you selected.

For regular scenarios, we show rates of return per asset, and how it changes over time if you create any overrides. For randomized scenarios, we show asset allocations per asset, and how it changes over time if you create any overrides.

For complete transparency, we show the randomized data used for the projection as a chart that illustrates hypothetical growth of $1,000 during the time period of the projection, which is based on the selected randomized sequence that has been applied to the plan. Combined with the information above, this gives you and your client everything you need to understand the assumptions that were made for the randomized scenario.

Additional notes

• Returns are nominal, before taxes, calculated in Canadian dollars, and include dividends.
• Standard deviations and correlations are calculated based on historical data from 1990. We use the same indexes as the FP Canada Guidelines:
  • Inflation is based on the Bank of Canada's posted CPI.
  • Cash is based on the historical performance of FTSE 91-day T-Bill Index.
  • Fixed income is based on the historical performance of FTSE Universe Bond Index.
  • Equity is based on the historical performance of 1/3 of S&P/TSX Composite Index + 1/6 of S&P 500 Index + 1/6 of MSCI EAFE Index + 1/3 of MSCI Emerging Markets Index.
    • This combination of indexes represents the arithmetic average (equal weight) of the following regions used by the guidelines: Canada, Developed Markets, and Emerging Markets.
    • We use the same weights to calculate the default rate of return for equities when you choose to use the guidelines for a new scenario.
• Return and inflation assumptions are assumed to be geometric means. These rates are converted to arithmetic means using an approximation formula of arithmetic mean = geometric mean + variance/2. The resulting arithmetic means are used to generate the randomized sequences that you can select from in the randomized modal.
• If an Expected standard deviation is set to 0%, the actual value used by Snap is 0.001 to avoid dividing by 0 in formulas used throughout this feature.
• Monthly returns have been approximated from annual returns.
• Past returns are not indicative of future returns.

Back to top

Activation

If the Stress Testing add-on has already been activated, you'll be able to access the Randomized Scenarios feature right away. Otherwise, you will be prompted to activate the add-on when you try to access it for the first time. If you click Activate & Update Billing, you will activate the add-on only for yourself. Other users in your organization will not be affected.

If you are the administrator of your organization (or you have an individual account), you can also activate or deactivate this feature for all users in your organization on the Billing Settings page.

Deactivation

If you deactivate this feature, your existing Randomized Scenarios will continue to be based on the previously selected randomized data, the assumptions will remain the same, and you will be able to rerun your existing randomized scenarios, but you will lose access to the randomized modal and won't be able to make any further adjustments to the rates unless you reactivate the add-on.

Even when the feature is inactive, you still can convert your existing Randomized Scenarios back to regular scenarios. Just go to the Assets page of your scenario and click Clear Randomized Returns. See clearing randomized data for more information.

Back to top