Medcn: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Jeremy
No edit summary
imported>Scott
 
(4 intermediate revisions by 2 users not shown)
Line 11: Line 11:
MEDCN centers a matrix x to it's median and returns a matrix mcx with median zero columns and a vector of medians mx used to center the data. Optional input options is discussed below.
MEDCN centers a matrix x to it's median and returns a matrix mcx with median zero columns and a vector of medians mx used to center the data. Optional input options is discussed below.


The output msg returns any warning messages.
The output <tt>msg</tt> returns any warning messages.


===Options ===
===Options ===
Line 18: Line 18:


* '''display''': [ {'off'} | 'on'] Governs screen display.
* '''display''': [ {'off'} | 'on'] Governs screen display.
* '''matrix_threshold''': {.15} Error threshold based on fraction of missing data in whole matrix.
* '''matrix_threshold''': [.15] Error threshold based on fraction of missing data in whole matrix.
* '''column_threshold''': {.25} Error threshold based on fraction of missing data in single column.
* '''column_threshold''': [.25] Error threshold based on fraction of missing data in single column.


===See Also===
===See Also===


[[auto]], [[mncn]], [[percentile]], [[rescale]], [[scale]]
[[auto]], [[mncn]], [[percentile]], [[polytransform]], [[rescale]], [[scale]]

Latest revision as of 10:07, 21 September 2011

Purpose

Median center scales matrix to median zero.

Synopsis

[mcx,mx,msg] = medcn(x,options)

Description

MEDCN centers a matrix x to it's median and returns a matrix mcx with median zero columns and a vector of medians mx used to center the data. Optional input options is discussed below.

The output msg returns any warning messages.

Options

options = a structure array with the following fields.

  • display: [ {'off'} | 'on'] Governs screen display.
  • matrix_threshold: [.15] Error threshold based on fraction of missing data in whole matrix.
  • column_threshold: [.25] Error threshold based on fraction of missing data in single column.

See Also

auto, mncn, percentile, polytransform, rescale, scale