Plot_Partial_RSS_OnLine
Location: ..\cubatch\Model_nPLS1\private\Plot_Partial_RSS_OnLine.m
function
[Axes,MO] = Plot_Partial_RSS_OnLine (PlotStruct,ModelOut,MainFig);
Description:
It plots the sum of squares of the residuals (Q statistics)
computed in an online fashion or the Squared Prediction Error (SPE) vs time
(scalars in the last mode, which has to be given name:
'time'). The
distiction between Q and SPE case is made by the
'userdata'
of the calling menu ('Q' for Q and
'SPE' otherwise).
The confidence limits, computed by means of
OnLineRes are automatically plot. If they where not originally present they
are computed and saved in ModelOut.model.stats.qlim and in
ModelOut.model.stats.spelim.
When the model has been applied to new data this plot automatically applies to
the new data; the confidence limits are computed on the original
data set.
This plot is currently available only for the X matrix.
For the axes specified by Axes the
'ModelOut'
application-defined data is set, and it contains a copy of ModelOut.(employed
by nPLS1DisplayInfo)
Inputs:
PlotStruct:
PlotStruct structure
ModelOut: ModelOut
structure for the model selected via the menus in the main window.
MainFig: main figure handle
Outputs:
Axes: handle of the axes where
Q or SPE
are plot.
MO: ModelOut structure with the updated statistic. Empty if no new statistic was computed.
Called by:
Model_nPLS1\nPLS1ActivatePlots
NB. The
actual
call is made from
nPLS1Plot via feval
and not from
Model_nPLS1\nPLS1ActivatePlots.
Subroutines:
Internal: None
External:
nplspred,
onlineres,
selectmodel
Author:
Giorgio Tomasi
Royal Agricultural and Veterinary University
MLI, LMT, Chemometrics group
Rolighedsvej 30
DK-1958 Frederiksberg C
Danmark
Last modified: 11-Oct-2002 18:46:22
Contact: Giorgio Tomasi, gt@kvl.dk
References
[1] - Jackson, J.E., Mudholkar, G.S., "Control Procedures for Residuals associated
with Principal Components Analysis", Technometrics, 21, (1979), 341
[2] - Nomikos P., Mac Gregor J.F., "Monitoring batch process using
multiway principal component analysis", AIChE journal, Vol 40, n°8, 1994,
1361-1373
[3] - Nomikos P., Mac Gregor J.F., "Multi-way partial least
squares in monitoring batch process", Chemometrics Intell. Lab Syst., Vol 30,
1995, 97-108