Downside deviation
The standard deviation of a fund's negative-return periods only. Input to the Sortino ratio. Captures asymmetric downside that total standard deviation hides.
Downside deviation is a one-sided variant of standard deviation that includes only periods where the fund's return was below a chosen threshold (usually zero or a near-cash benchmark rate). Periods of upside are ignored.
Downside deviation is the input to the Sortino ratio and is also useful as a standalone measure for strategies with asymmetric return distributions. A fund with steady small gains and occasional sharp losses will have downside deviation close to its full σ; a fund with large positive surprises and small steady losses will have downside deviation well below its full σ.
NZ retail PIE fact sheets occasionally quote downside deviation for defensive and capital-protected mandates. The FMA risk indicator does not use downside deviation — the standardised risk metric is symmetric standard deviation.
Related terms
-
volatility · σ
Standard deviation (volatility)
The statistical measure of how widely a fund's returns vary around their average. The input to the FMA risk indicator: weekly returns over five years, mapped to a 1–7 band.
-
sortino-ratio
Sortino ratio
A downside-only variant of the Sharpe ratio: excess return divided by the standard deviation of negative returns only. Penalises losses but not symmetric upside volatility.
-
max DD · drawdown
Maximum drawdown
The largest peak-to-trough percentage decline in a fund's unit price over a measurement window. Captures worst-case past loss in a way standard deviation does not.