Difference between revisions of "Madc"

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
(Created page with "===Description=== The '''madc''' function is a scale estimator given by the Median Absolute Deviation (with finite sample correction factor). It is defined as: mad(x)= b_n 1...")
 
 
Line 2: Line 2:
 
The '''madc''' function is a scale estimator given by the Median Absolute Deviation (with finite sample correction factor).
 
The '''madc''' function is a scale estimator given by the Median Absolute Deviation (with finite sample correction factor).
 
It is defined as:  
 
It is defined as:  
  mad(x)= b_n 1.4826 med(|x_i - med(x)|)
+
  madc(x)= b_n 1.4826 med(|x_i - med(x)|)
 
with <code>b_n</code> a small sample correction factor to make the mad unbiased at the normal distribution. It can resist 50% outliers.  If <code>x</code> is a matrix, the scale estimate is computed on the columns of <code>x</code>. The result is then a row vector. If <code>x</code> is a row or a column vector, the output is a scalar.
 
with <code>b_n</code> a small sample correction factor to make the mad unbiased at the normal distribution. It can resist 50% outliers.  If <code>x</code> is a matrix, the scale estimate is computed on the columns of <code>x</code>. The result is then a row vector. If <code>x</code> is a row or a column vector, the output is a scalar.
  

Latest revision as of 10:48, 5 December 2019

Description

The madc function is a scale estimator given by the Median Absolute Deviation (with finite sample correction factor). It is defined as:

madc(x)= b_n 1.4826 med(|x_i - med(x)|)

with b_n a small sample correction factor to make the mad unbiased at the normal distribution. It can resist 50% outliers. If x is a matrix, the scale estimate is computed on the columns of x. The result is then a row vector. If x is a row or a column vector, the output is a scalar.

This function is part of LIBRA: the Matlab Library for Robust Analysis, available at:
http://wis.kuleuven.be/stat/robust.html
Written by S.Verboven

Synopsis

result = madc(x);

Inputs

  • x either a data matrix with n observations in rows, p variables in columns or a column vector of length n.

See Also

auto, box_filter, datafit_engine, windowfilter