Tsqqmtx: Difference between revisions
Jump to navigation
Jump to search
imported>Jeremy (New page: ===Purpose=== Calculates matrix for T^2+Q contributions for PCA and MPCA. ===Synopsis=== :[tsqqmat,tsqqs] = tsqqmtx(x,model,wt) ===Description=== ====Inputs==== * '''x = data matrix [...) |
imported>Scott |
||
| (One intermediate revision by one other user not shown) | |||
| Line 10: | Line 10: | ||
====Inputs==== | ====Inputs==== | ||
* '''x = data matrix [class double or dataset] | * '''x''' = data matrix [class double or dataset] | ||
* '''model = PCA or MPCA model standard model struture (see PCA). | * '''model''' = PCA or MPCA model standard model struture (see PCA). | ||
====Optional Inputs==== | ====Optional Inputs==== | ||
| Line 25: | Line 25: | ||
===See Also=== | ===See Also=== | ||
[[datahat]], [[pca]], [[pcr]], [[pls]], [[tsqmtx]] | [[datahat]], [[pca]], [[pcr]], [[pls]], [[tconcalc]], [[tsqmtx]] | ||
Latest revision as of 16:05, 21 September 2011
Purpose
Calculates matrix for T^2+Q contributions for PCA and MPCA.
Synopsis
- [tsqqmat,tsqqs] = tsqqmtx(x,model,wt)
Description
Inputs
- x = data matrix [class double or dataset]
- model = PCA or MPCA model standard model struture (see PCA).
Optional Inputs
- wt = {sqrt((M-K-1)/(M-1))}, 0<=wt<=1 scalar weighting for contributions 0<wt<1 gives combined T^2 and Q statistics where M is the number of calibration samples and K is the number of PCs.
- wt = 1 gives T^2 and T^2 contributions
- wt = 0 gives standarized Q residuals
Outputs
- tsqqs = combined Hotelling's T^2 + Q residual
- tsqqmat = matrix of individual variable contributions such that
- tsqqs(i) = tsqqmat(i,:)*tsqqmat(i,:)';