User Guide

• Understanding the inputs, assumptions, and settings that drive the model

• Properly interpreting the graph and data table for simulation results

• Getting the most out of the analysis and avoiding common mistakes

Technology

State-of-the-art architecture

We’ve partnered with Amazon Web Services (AWS) to leverage the power, speed, and elasticity of the world’s largest cloud-computing platform. This allows us to cost-efficiently execute the tens of millions of calculations required for each simulation in a matter of seconds.

Portfolio

Specify your investable assets, savings habits, and investing style

Portfolio Balance
How much do you have saved for retirement?

Asset Allocation
How is your money allocated between stocks and bonds? Add rows to adjust your holdings over time (e.g., shift from stocks to bonds as you age), if you’d like.

Regular Contributions
If you’re not currently retired, tell us how much of your income you’re saving for retirement each month. You can express this as a percentage of gross income or as a dollar amount.

Special Contributions
If you anticipate occasional irregular portfolio contributions or one-time increases to your nest egg (e.g., selling your primary residence or inheriting wealth), specify it here.

Settings

Modify capital market assumptions, calibrate the statistical parameters, or specify a Black Swan event

Capital Market Assumptions
Our default assumptions are informed by the prevailing long-term capital market views of prominent investment banks. These parameters are forward-looking, and will change over time as market conditions evolve. Users can modify these assumptions at their discretion. However, we advise caution when deviating from the default inputs, and strongly discourage the use of historical averages.

• Riskier Assets (“Stocks”)
  The first column (labeled “Stocks”) is designed to model riskier investments. The asset entered into this column is simulated with a fat-tailed probability distribution, which can be calibrated or switched off in the “Fat Tails” section. The default assumptions reflect long-term expectations for U.S. large cap equities (e.g., S&P 500 Index). 

• Less Risky Assets (“Bonds”)
  The second column (labeled “Bonds”) is designed to model less risky investments. The asset entered into this column is simulated with a Gaussian distribution (e.g., normal bell curve). The default assumptions for this column reflect the long-term expectations for U.S. investment grade corporate fixed-income securities (e.g., Bloomberg Barclays U.S. Corporate Bond Index).

• Expected Return: Expected return should be entered as a nominal annual average on an arithmetic basis, not a geometric basis. The arithmetic mean is the baseline location parameter for the random sampling process. The geometric mean is the effective annual compound return, such as the rates of return reported on the results page. In other words, the arithmetic mean is the input, whereas the geometric mean is the output. We explore the concept of expected return is greater detail here.

• Standard Deviation: Enter standard deviation on an annualized basis to dictate the expected volatility of each asset class. The volatility for stocks can be further modified by calibrating the “Fat Tails” settings. We explain standard deviation in greater detail here.

• Correlation Coefficient: Enter the correlation coefficient that describes the linear relationship between the two portfolio assets. This value should fall anywhere from -1.00 to 1.00, and it helps the simulator appropriately model portfolio behavior. We explain the correlation coefficient in greater detail here.

Fat Tails
Equity markets are more volatile than indicated by statistical models traditionally used to simulate investment performance (e.g., the normal bell curve). By default, our equity performance model uses a stochastic method called a levy process to capture the excess kurtosis of stock prices observed empirically. In other words, we “fatten” the tails of the sampling distribution to more realistically replicate market volatility. Our methodology is the culmination of extensive research and testing. Although the algorithms for implementing our stochastic process are proprietary, the mathematical principals underpinning the framework are well established in academic literature.

 Magnitude of Tail Events 

• Standard: Tail event magnitude is consistent with historical precedent in U.S. markets.

• Extreme: Tail event magnitude exceeds historical precedent in U.S. markets.

Frequency of Tail Events 

• Standard: Tail event frequency is consistent with historical precedent in U.S. markets.

• Extreme: Tail event frequency is doubled relative to historical precedent in U.S. markets.

Skewness of Tail Events

• Negative: Most tail events are losses (left tail occurrences).

• Neutral: Tail event distribution approximates historical precedent in U.S. markets.

• Positive: Most tail events are gains (right tail occurrences). It’s fun to dream, isn’t it? Please use with appropriate caution, my optimistic friends.

Black Swan Event
Markets occasionally suffer sudden and extreme shocks, often referred to as tail events or a “Black Swan”. This term was popularized by Nassim Taleb, a statistician, risk analyst, and former options trader that authored a book by the same name. The Black Swan feature allows users to stress test their retirement plan by applying a sudden and extreme drop in portfolio value to every simulation trial.