How to Calculate Portfolio Standard Deviation and Volatility Risk

Key Takeaways

The S&P 500's long-run annualized standard deviation is approximately 15.1%. A portfolio with 22% standard deviation takes on 46% more volatility for every unit of return.
Investors who ignore correlation between assets often overestimate diversification, leaving portfolios exposed to $40,000+ drawdowns on a $200,000 balance during moderate corrections.
Calculate portfolio standard deviation using each asset's weight, individual standard deviation, and pairwise correlations, not a simple average of individual volatilities.
Tool: Run your portfolio volatility numbers now →

Invest Your Spare Change Automatically

Acorns rounds up your purchases and invests the difference into a diversified portfolio automatically.

Interactive Calculator

Full screen

Loading Calculator

calcmoney.io/calculatorsOpen full screen

Standard Deviation Measures How Much Your Returns Deviate From Their Own Average

Portfolio standard deviation quantifies how widely a portfolio's periodic returns scatter around their mean. A portfolio with a 10% average annual return and a 5% standard deviation will produce returns between 5% and 15% roughly 68% of the time. Widen that standard deviation to 18%, and that same 68% confidence band stretches from -8% to +28%. The expected return looks identical. The risk does not.

The formula for a single asset's standard deviation uses historical return observations:

Standard Deviation = Square Root of [ (Sum of (Each Return - Mean Return)^2 ) / (Number of Observations - 1) ]

For a portfolio, the calculation becomes more involved because individual asset volatilities interact through correlation.

The Two-Asset Portfolio Formula Introduces Correlation

A two-asset portfolio's standard deviation is not a weighted average of the two individual standard deviations. Correlation between the assets reduces, or in rare cases increases, total volatility.

The formula for a two-asset portfolio:

Portfolio Std Dev = Square Root of [ (W1^2 x Std1^2) + (W2^2 x Std2^2) + (2 x W1 x W2 x Std1 x Std2 x Correlation) ]

Where W1 and W2 are the portfolio weights, Std1 and Std2 are the individual standard deviations, and Correlation is the Pearson correlation coefficient between the two assets' return series.

Worked Example 1: A $250,000 Two-Asset Portfolio

An investor holds $175,000 in a broad US equity index fund and $75,000 in a 10-year Treasury bond fund. That is a 70/30 split.

W1 (equities) = 0.70, Std1 = 17.2%
W2 (bonds) = 0.30, Std2 = 6.8%
Correlation between the two = -0.25 (historically typical for stocks and investment-grade bonds)

Step 1. Square each weighted standard deviation:

(0.70)^2 x (0.172)^2 = 0.49 x 0.02958 = 0.01449
(0.30)^2 x (0.068)^2 = 0.09 x 0.004624 = 0.000416

Step 2. Calculate the cross term:

2 x 0.70 x 0.30 x 0.172 x 0.068 x (-0.25) = -0.001229

Step 3. Sum and take the square root:

0.01449 + 0.000416 + (-0.001229) = 0.013677
Square root of 0.013677 = 0.1170, or 11.70%

A naive average of the two standard deviations weighted by allocation would produce 13.1%. The negative correlation shaves 1.4 percentage points off realized volatility. On a $250,000 portfolio, that difference translates to roughly $3,500 in reduced expected annual swing at one standard deviation.

Extending the Formula to Three or More Assets Requires Matrix Math

For portfolios with three or more assets, the standard deviation calculation uses a covariance matrix. Each cell in the matrix holds the covariance between a pair of assets. Covariance equals the correlation between two assets multiplied by each asset's standard deviation.

Covariance(A, B) = Correlation(A, B) x Std(A) x Std(B)

The portfolio variance is then calculated as the weighted sum of all entries in the covariance matrix, where weights appear in both the row and column dimensions. The portfolio standard deviation is the square root of that variance.

Most investors use a spreadsheet or a purpose-built calculator for three-asset or larger portfolios. The underlying logic remains the same. Correlation is doing most of the work.

Worked Example 2: Adding a Third Asset Changes the Outcome Materially

Take the same $250,000 portfolio from Example 1. The investor replaces $50,000 of the equity position with a real estate investment trust fund, leaving $125,000 in equities, $75,000 in bonds, and $50,000 in REITs.

W1 = 0.50 (equities, Std = 17.2%)
W2 = 0.30 (bonds, Std = 6.8%)
W3 = 0.20 (REITs, Std = 19.4%)
Correlation, equities/bonds = -0.25
Correlation, equities/REITs = 0.62
Correlation, bonds/REITs = 0.10

Variance components:

(0.50)^2 x (0.172)^2 = 0.25 x 0.02958 = 0.007396
(0.30)^2 x (0.068)^2 = 0.09 x 0.004624 = 0.000416
(0.20)^2 x (0.194)^2 = 0.04 x 0.037636 = 0.001505
2 x 0.50 x 0.30 x 0.172 x 0.068 x (-0.25) = -0.000879
2 x 0.50 x 0.20 x 0.172 x 0.194 x 0.62 = 0.004149
2 x 0.30 x 0.20 x 0.068 x 0.194 x 0.10 = 0.000158

Sum of all components: 0.007396 + 0.000416 + 0.001505 - 0.000879 + 0.004149 + 0.000158 = 0.012745

Square root of 0.012745 = 0.1129, or 11.29%

Adding REITs at a 0.62 correlation to equities actually reduced total portfolio standard deviation slightly, from 11.70% to 11.29%, because the bonds' negative correlation with both other assets offset the high REIT/equity correlation. The $250,000 portfolio's one-standard-deviation range narrowed by roughly $1,025 per year in expected swing.

Annualizing Standard Deviation From Monthly or Weekly Data

Most return data arrives monthly or weekly. Annualizing requires multiplying the periodic standard deviation by the square root of the number of periods in a year.

Monthly standard deviation x Square Root of 12 = Annualized standard deviation
Weekly standard deviation x Square Root of 52 = Annualized standard deviation

A portfolio showing a monthly standard deviation of 3.38% annualizes to 3.38% x 3.464 = 11.70%. Skipping this step causes investors to dramatically understate risk when comparing portfolios to annual benchmarks like the S&P 500's 15.1% historical figure.

What Standard Deviation Does Not Capture

Standard deviation assumes returns follow a roughly normal distribution. Real return distributions show fat tails, meaning extreme losses occur more frequently than a normal distribution predicts. The 2008 financial crisis produced a drawdown of 56.8% in the S&P 500, a move that standard deviation models based on prior calm periods did not adequately forecast.

Standard deviation also treats upside and downside volatility identically. An investor with a $500,000 portfolio approaching retirement cares more about downside deviation than total volatility. Sortino ratio and downside deviation address this distinction directly, using only negative return periods in the calculation.

Use standard deviation as a relative comparison tool, not an absolute safety measure.

Use the CalcMoney Investment Calculator to Run These Numbers on Your Actual Portfolio

Manual covariance matrix calculations for a five-asset or ten-asset portfolio involve dozens of pairwise computations. One transposed weight or misplaced decimal corrupts the entire result.

The CalcMoney investment calculator runs portfolio standard deviation, annualized volatility, and risk-adjusted return metrics using your actual allocation inputs. Enter your weights and asset class assumptions, and the calculator returns the full variance decomposition in seconds.

Investors managing $100,000 or more should not rely on gut feel about diversification. The math either supports the allocation or it does not.

Calculate your portfolio's standard deviation now →

Results are estimates for informational purposes only. Consult a licensed financial professional before making financial decisions.

Featured Partner

FIDELITY