Tsqlim and Release Notes MIA Toolbox Version 3 0 5: Difference between pages

From Eigenvector Research Documentation Wiki
(Difference between pages)
Jump to navigation Jump to search
imported>Scott
 
imported>Scott
No edit summary
 
Line 1: Line 1:
===Purpose===
==Changes and Bug Fixes in Version 3.0.5==


Calculates PCA confidence limits for Hotelling's T<sup>2</sup>.
===Bug Fixes and Enhancements===
 
{|
===Synopsis===
|----valign="top"
 
|'''[[box_filter]]'''
:tsqcl = tsqlim(m,pc,cl)
|
:tsqcl = tsqlim(model,cl)
* Update help, added new options, and changed despike std calc from window to median of the window.
 
|----valign="top"
===Description===
|'''[[cluster_img]]'''
 
|
Inputs can be in one of two forms:
* Add option to plot preprocessed data.
 
|----valign="top"
(a) the number of samples <tt>m</tt>, the number of principal components used <tt>pc</tt>, and the fractional confidence limit, <tt>cl</tt> (0 < cl < 1) which can be a scalar or a vector (to calculate multiple confidence limits simultaneously).
|'''hidepixels'''
 
|
or (b) a standard model structure, <tt>model</tt>, and the fractional confidence limit, <tt>cl</tt> (0 < cl < 1).
* Standalone utility to hidepixels.
 
|----valign="top"
The output <tt>tsqcl</tt> is the confidence limit based on an F distribution as shown below. See Jackson (1991).
|'''coadd_imgdemo'''
 
|
    <math>T_{K,M,\alpha }^{2}=\frac{K\left( M-1 \right)}{M-K}{{F}_{K,M-K,\alpha }}</math>
* Add demo script for coadd_img.
 
|----
where <math>K</math> is the number of PCs, <math>M</math> is the number of samples and <math>{{F}_{K,M-K,\alpha }}</math> is the F distribution with <math>K</math> degrees of freedom in the numberator and <math>M-K</math> degrees of freedom in the denominator, and probability point <math>\alpha</math>.
|}
 
===Examples===
 
tsqcl = tsqlim(15,2,0.95)
 
model = pca(data,pc); tsqcl = tsqlim(model,0.95)
 
===See Also===
 
[[analysis]], [[chilimit]], [[pca]], [[pcr]], [[pls]], [[subgroupcl]]

Latest revision as of 12:53, 18 January 2018

Changes and Bug Fixes in Version 3.0.5

Bug Fixes and Enhancements

box_filter
  • Update help, added new options, and changed despike std calc from window to median of the window.
cluster_img
  • Add option to plot preprocessed data.
hidepixels
  • Standalone utility to hidepixels.
coadd_imgdemo
  • Add demo script for coadd_img.