Wikia

Psychology Wiki

Median absolute deviation

Talk0
34,136pages on
this wiki
Revision as of 01:15, November 4, 2007 by AWeidman (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

In statistics, the median absolute deviation (or "MAD") is a resistant measure of the variability of a univariate sample. It is useful for describing the variability of data with outliers.

For a univariate data set X1X2, ..., Xn, the MAD is defined as


\operatorname{MAD} = \operatorname{median}_{i}\left(\ \left| X_{i} - \operatorname{median}_{j} (X_{j}) \right|\ \right), \,

that is, starting with the residuals (deviations) from the data's median, the MAD is the median of their absolute values.

Relation to standard deviation Edit

As an estimate for the standard deviation σ, one takes

\hat{\sigma}=K\cdot \operatorname{MAD},

where K is a constant. For normally distributed data K is taken to be 1 / Φ-1(3/4) (where Φ-1 is the inverse of the cumulative distribution function for the standard normal distribution), or 1.4826... , because the MAD is given by:

\frac 12 =P(|X-\mu|\le \operatorname{MAD})=P\left(\left|\frac{X-\mu}{\sigma}\right|\le \frac {\operatorname{MAD}}\sigma\right)=P\left(|Z|\le \frac {\operatorname{MAD}}\sigma\right).

Hence

\frac {\operatorname{MAD}}\sigma=\Phi^{-1}(3/4) \approx 0.6745

and:

\sigma \approx 1.4826\ \operatorname{MAD}.


In this case, its expectation for large samples of normally distributed Xi is approximately equal to the standard deviation of the normal distribution.

ReferencesEdit

  • Hoaglin, David C.; Frederick Mosteller and John W. Tukey (1983). Understanding Robust and Exploratory Data Analysis, 404-414, John Wiley & Sons.
  • Venables, W.N.; B.D. Ripley (1999). Modern Applied Statistics with S-PLUS, 128, Springer.

Around Wikia's network

Random Wiki