Varcap: Difference between revisions
Jump to navigation
Jump to search
imported>Jeremy (Importing text file) |
imported>Jeremy No edit summary |
||
| Line 1: | Line 1: | ||
===Purpose=== | ===Purpose=== | ||
| Line 12: | Line 11: | ||
VARCAP calculates and displays the percent variance captured for each variable and number of principal components in a PCA model. | VARCAP calculates and displays the percent variance captured for each variable and number of principal components in a PCA model. | ||
Inputs are the properly scaled ''M'' by ''N'' data x (''i.e.'' scaled using the same scaling used when creating the PCA model) with associated ''N'' by ''K'' loadings matrix loads. | Inputs are the properly scaled ''M'' by ''N'' data <tt>x</tt> (''i.e.'' scaled using the same scaling used when creating the PCA model) with associated ''N'' by ''K'' loadings matrix <tt>loads</tt>. | ||
Optional input ''scl'' (1 by ''N'') specifies the x-axis for plotting. Optional input ''plots'' suppresses plotting when set to 0 {default = 1}. | Optional input ''<tt>scl</tt>'' (1 by ''N'') specifies the x-axis for plotting. Optional input ''<tt>plots</tt>'' suppresses plotting when set to 0 {default = 1}. | ||
The output is a ''K'' by ''N'' matrix of variance captured vc for each variable and each number of PCs considered (vc is number of PCs by number of variables). A stacked bar chart of vc is also plotted | The output is a ''K'' by ''N'' matrix of variance captured <tt>vc</tt> for each variable and each number of PCs considered (<tt>vc</tt> is size: [number of PCs by number of variables]). A stacked bar chart of <tt>vc</tt> is also plotted (See optional input ''<tt>plots</tt>''). | ||
===See Also=== | ===See Also=== | ||
[[analysis]], [[pca]] | [[analysis]], [[pca]] | ||
Revision as of 10:56, 10 October 2008
Purpose
Variance captured for each variable in PCA model.
Synopsis
- vc = varcap(x,loads,scl,plots)
Description
VARCAP calculates and displays the percent variance captured for each variable and number of principal components in a PCA model.
Inputs are the properly scaled M by N data x (i.e. scaled using the same scaling used when creating the PCA model) with associated N by K loadings matrix loads.
Optional input scl (1 by N) specifies the x-axis for plotting. Optional input plots suppresses plotting when set to 0 {default = 1}.
The output is a K by N matrix of variance captured vc for each variable and each number of PCs considered (vc is size: [number of PCs by number of variables]). A stacked bar chart of vc is also plotted (See optional input plots).