# Standard Deviation (STDDEV)

> StratForge indicator page for rolling dispersion of the input series.

**Route**: `/indicators/stddev/`

## What It Does

`stratforge::StdDev` computes the **population standard deviation** of the last `period` input values. Output has the same unit as the input price; it is not a percentage. The implementation calculates the window mean and population variance, then takes the square root.

## Formula and worked example

`StdDev = sqrt(sum((Price - mean)^2) / period)`

For `period = 4` and prices `[2, 4, 6, 8]`, the mean is `5`, squared deviations are `[9, 1, 1, 9]`, and variance is `20 / 4 = 5`. The output is `sqrt(5) ≈ 2.236` price units. A sample standard deviation would instead divide by `3`; that is **not** this indicator.

The output is unavailable until four bars exist. A one-bar window yields zero. The current implementation normalizes `period = 0` to a one-bar window; pass a positive period to make intent clear.

## Parameters

- `period` — rolling window size; required by the constructor.

## C++23 API

```cpp
#include <stratforge/indicators/stddev.hpp>
auto stddev = std::make_unique<stratforge::StdDev>(data().close(), 20);
```

The example assumes a StratForge strategy with `data()` available and `<memory>` included. [Checked implementation](https://github.com/StratCraftsAI/StratForge/blob/d25404dcf2c3842ba304ddc60ac19cfec1f8aa37/include/stratforge/indicators/stddev.hpp) (revision `d25404d`, 2026-08-28).

## Common Usage

- Use price-unit dispersion in volatility filters or bands.
- To compare assets at different price levels, normalize by a defined price baseline and handle zero explicitly.
- A larger standard deviation describes spread in this window, not future returns.