FTB-C000431 / Statistical method

Mean Absolute Deviation

Mean Absolute Deviation is the arithmetic mean of absolute distances from a declared center.

Also known asmean deviationMAD from mean

Definitions

In plain terms

CCI uses the window typical-price mean as the center and the same window observations for every distance.

Technical

The contract freezes center, population, denominator, window, current-value inclusion, missing values, and zero-deviation policy.

Scope

It is not median absolute deviation and must not be replaced by standard deviation.

Formula

Mean deviation = sum(abs(x_i - mean(x))) / n
LaTeX: MD=\frac{1}{n}\sum_i|x_i-\bar{x}|
SymbolMeaningUnit
x_iwindow observation iinput unit
mean(x)arithmetic mean of the same observationsinput unit
nnumber of observationsobservations

Output unit: input unit

Examples

  • A governed lesson calculates or identifies Mean Absolute Deviation only after its inputs, window, state, scale, and edge cases are declared.

Common misconceptions

  • It is not median absolute deviation and must not be replaced by standard deviation.

Concept relationships

Where this concept is used

Evidence and governance

  1. Mean Absolute Deviation NIST · official standard

    Supports: preferred label, short definition, technical definition, formula

    Limits: CCI specifically uses the mean of absolute deviations of typical prices from their window arithmetic mean; other centers are different measures.

  2. Commodity Channel Index TA-Lib · first party technical publication

    Supports: variant distinction

    Limits: The common 0.015 constant is a convention, and zero-deviation or mixed-adjustment inputs require explicit handling.

Reviewed by
fintech-builder-batch-008
Last reviewed
2026-07-29
Next review
2027-07-29
Record status
published