NISM Professor

Standard deviation

Also written Sigma

A measure of how far returns typically stray from their own average — the standard statistic for total risk, counting company-specific and market-wide causes alike.

In plain language

Two funds both averaged 12% a year over three years. The first returned 11%, 12% and 13%. The second returned −8%, 12% and 32%.

The average is identical and the experience of owning them is not remotely comparable. Standard deviation is the number that tells them apart: it measures the typical distance between a year's return and the average return.

How it works

Take each year's deviation from the mean, square it so that overshoots and undershoots do not cancel, average the squares to get the variance, then take the square root to return to percentage units.

If returns are roughly normally distributed, the result reads directly as a range: about two years in three fall within one standard deviation of the mean, and about nineteen in twenty within two. A fund averaging 12% with a standard deviation of 20% should therefore be expected to land between −8% and +32% in two years out of three — and worse than −28% in one year in twenty.

The formula

σ = √[ Σ (Rᵢ − R̄)² ÷ (n − 1) ]

Variance = σ²

Divide by n − 1 for a sample and by n for a full population. NISM questions almost always intend the sample form.

A worked example

Two equity funds, five years of annual returns, both averaging exactly 12%:

YearFund AFund B
114%38%
29%−14%
312%26%
411%4%
514%6%
Mean12%12%
Fund A: deviations  2, −3, 0, −1, 2
        squares     4,  9, 0,  1, 4  = 18
        variance    18 ÷ 4 = 4.5      σ = 2.1%

Fund B: deviations 26, −26, 14, −8, −6
        squares   676, 676, 196, 64, 36 = 1,648
        variance  1,648 ÷ 4 = 412       σ = 20.3%

Same average return, ten times the volatility. On Rs 10 lakh, Fund A's worst year still returned Rs 90,000; Fund B's worst year lost Rs 1.4 lakh. An investor who needed the money in year two experienced two completely different products.

Why NISM asks about it

Chapter 12 (Fundamentals of Risk and Return) introduces standard deviation as the measure of total risk, and it returns as the denominator of the Sharpe ratio. Expect a short series of returns to compute from, and questions that turn on the difference between variance and standard deviation.

Common exam traps

  • Variance is in squared units and standard deviation is its square root. Answer options routinely offer both; only one is comparable to a return.
  • Standard deviation is total risk, beta is systematic risk. A question asking for the risk that diversification cannot remove is not asking for this.
  • It treats upside and downside identically. A fund whose surprises are all pleasant still scores as risky.
  • A portfolio's standard deviation is not the weighted average of its holdings'. Correlation below 1 makes the portfolio less volatile than its parts — which is the mathematics of diversification.
  • Comparing standard deviations computed over different periods, or on monthly against annual data, is meaningless.

Where this is taught

Free preparation for NISM Series XIX-B

Related terms

← All terms
Something look wrong? Report it