TAlib.Indicators.StdDev (TAlib v0.13.0)

Copy Markdown View Source

Rolling population standard deviation.

See TAlib for the input conventions: prices are always given in chronological order, oldest first, and the series is returned aligned with the input. nil is returned when there is not enough data.

Summary

Functions

Warm up streaming standard-deviation state from history.

Advance streaming standard-deviation state by one price, in O(1).

Calculate the population standard deviation of the period newest prices (two-pass formula, dividing by period).

Calculate the full rolling standard deviation series in a single pass (sliding window sum and sum of squares).

Functions

init(prices, period \\ 50)

@spec init([number()], pos_integer()) ::
  {TAlib.Indicators.StdDev.State.t(), [float() | nil]}

Warm up streaming standard-deviation state from history.

Returns {state, values} with values the full series of history (identical to std_dev_list/2), and state ready to continue it one price at a time in O(1) via next/2. History needs at least period prices.

Examples

iex> {state, values} = TAlib.Indicators.StdDev.init([2, 4, 6, 8], 2)
iex> values
[nil, 1.0, 1.0, 1.0]
iex> {_state, 1.0} = TAlib.Indicators.StdDev.next(state, 10)

next(state, price)

@spec next(TAlib.Indicators.StdDev.State.t(), number()) ::
  {TAlib.Indicators.StdDev.State.t(), float() | nil}

Advance streaming standard-deviation state by one price, in O(1).

iex> {state, _} = TAlib.Indicators.StdDev.init([2, 4, 6, 8], 2)
iex> {_state, value} = TAlib.Indicators.StdDev.next(state, 10)
iex> value
1.0

std_dev(prices, period \\ 50)

@spec std_dev([number()], pos_integer()) :: float() | nil

Calculate the population standard deviation of the period newest prices (two-pass formula, dividing by period).

Parameters

  • prices: list of prices, oldest first.
  • period: lookback window, a positive integer.

Examples

iex> TAlib.Indicators.StdDev.std_dev([1, 2, 3, 4], 4) |> Float.round(6)
1.118034

iex> TAlib.Indicators.StdDev.std_dev([2, 4, 6, 8], 2)
1.0

std_dev_list(prices, period \\ 50)

@spec std_dev_list([number()], pos_integer()) :: [float() | nil]

Calculate the full rolling standard deviation series in a single pass (sliding window sum and sum of squares).

Returns a list aligned with prices: the first period - 1 entries are nil, and every later entry is the population standard deviation of the window ending at that point.

Parameters

  • prices: list of prices, oldest first.
  • period: lookback window, a positive integer.

Examples

iex> TAlib.Indicators.StdDev.std_dev_list([2, 4, 6, 8], 2)
[nil, 1.0, 1.0, 1.0]