archive-com.com » COM » E » EIGENVECTOR.COM Total: 295 Choose link from "Titles, links and description words view": Or switch to
"Titles and links view". |

length y vector w 1 selected 0 unselected a figure NumberTitle Off Name Spectra Edit plot f y b LineWidth 0 5 MarkerSize 6 hold on h1 plot f 1 1 y 1 1 r Visible Off hold off p 1 1 4 get a position p 1 1 p 1 3 63 p 1 2 p 1 4 114 p 1 3 4 0 0 b zeros 5 5 b 1 2 5 0 0 61 112 b 2 2 5 3 3 55 25 b 3 2 5 3 30 55 25 b 4 2 5 3 57 55 25 b 5 2 5 3 84 55 25 b 1 1 uicontrol Parent a BackgroundColor bgc1 Units pixels position p b 1 2 5 Style Frame UserData f b 2 1 uicontrol Parent a CallBack specedit savespc String save spec UserData x b 3 1 uicontrol Parent a CallBack specedit saveind String save inds UserData y b 4 1 uicontrol Parent a CallBack specedit deselect String deselect UserData h1 y b 5 1 uicontrol Parent a CallBack specedit select String select UserData ind0 for jj 2 5 set b jj 1 position p b jj 2 5 BackgroundColor bgc2 Interruptible off BusyAction cancel Units pixels FontName fnam Fontsize fsiz end set a UserData b zoompls s specedit SPECEDTsize zoompls ZOOMPLSsize set a ResizeFcn s else b get gcf UserData if strcmp action savespc strcmp action saveind indedt get b 5 1 UserData indedt find indedt if strcmp action savespc xedt get b 2 1 UserData xedt xedt indedt svdlgpls xedt else svdlgpls indedt end elseif strcmp action SPECEDTsize p 1 1 4 get gcf position p 1 1 p 1 3 63 p 1 2 p 1 4 114 p 1 3 4 0 0 for jj 1

Original URL path: http://eigenvector.com/MATLAB/Users_Mac/specedit.m (2016-04-27)

Open archived version from archive

wanted in the model set int 1 default value If an intercept is not wanted set int to 0 The X matrix is required to have samples in rows and variables in the columns OUTPUT The regression coefficents are returned by the function beta vector The model prediction information is also returned in the stats output The observed y predicted y relative percent error RPE standard error of the mean and standard error of a future observation at the given y are contained in stats in that order IMPORTANT NOTE To fit an MLR model the number of samples must be greater than the number of independent variables If this is not the case PLS or PCR models should be fit using MODLMAKER or MODLGUI or some variables should be eliminated As the number of variables with or without intercept gets closer to the number of samples the X X inverse matrix may become unstable and may not exist i e singular X X matrix Kirk Remund Pete Eschbach Pacific Northwest National Laboratory Battelle e mail km remund pnl gov pa eschbach pnl gov phone 509 372 4729 375 2678 pete fax 509 375 3614 372 4725 pete Disclaimer Neither of the authors nor the Battelle Memorial Institute claim any responsibility for the accuracy of results obtained using this function if nargin columns independent vars if p n error More variables than samples use less variables and or omit intercept end a b size y make y a column vector if a 10 print first 10 records z 10 else z n end for i 1 z tab sprintf format stats i disp tab end disp pause disp s sprintf Root Mean Square Error of Calibration is g rmsec disp s disp s sprintf Root Mean Square Error from ANOVA

Original URL path: http://eigenvector.com/MATLAB/Mac_M_files/mlr.m (2016-04-27)

Open archived version from archive

applications The input is the data matrix dat The output is the matrix of normalized data ndat and the vector of norms used in the normalization norms Warnings are given for any zero vectors found The I O syntax is ndat norms normaliz dat Copyright 1997 Eigenvector Research Inc Barry M Wise May 30 1997 m n size dat ndat dat norms zeros m 1 for i 1 m if

Original URL path: http://eigenvector.com/MATLAB/PC_M_files/normaliz.m (2016-04-27)

Open archived version from archive

by the loadings from PLS or PCR using the number of LVs or PCs in the corresponding calibration model The outputs are the matrix of net analyte signals for each of the spectra nas the norm of the net analyte signal for each sample nnas the matrix of sensitivities for each sample sens the vector of selectivities for each sample sel and the noise filtered estimate of the net analyte signal nfnas which is just the multiple of the regression vector that best fits the nas For example given the 7 LV PLS model formed from p q w t u b ssqdif pls x y 7 Rhat t p nas nnas sens sel nfnas figmerit x y Rhat Copyright 1997 Eigenvector Research Inc Barry M Wise May 30 1997 mx nx size x nas zeros mx nx nnas zeros mx 1 sel zeros mx 1 rhat mean Rhat find y 0 u s v svd Rhat 0 npcs max find diag s s 1 1 1e 10 Rhatinv u 1 npcs inv s 1 npcs 1 npcs v 1 npcs chatk Rhat Rhatinv y alpha inv rhat Rhatinv chatk Rhat k Rhat alpha chatk rhat u s v svd

Original URL path: http://eigenvector.com/MATLAB/Users_Mac/figmerit.m (2016-04-27)

Open archived version from archive

(No additional info available in detailed archive for this subpage)

Original URL path: /faq/preprouser_howto.html (2016-04-27)- Sampling Intro at CAC-2002

visit the top of the Space Needle Be sure to check out Pike Place Market and the Seattle Art Museum If you want general visitor information check out the Seattle Visitor Information Center or Seattle Sidewalk Introduction Sampling is an integral part of the analytical process and often takes the major part of the total uncertainty of the analytical measurement The importance of sampling is often recognized in textbooks and standards by phrases like sample must be representative or the result is not better than the sample They seldom give instructions however how the sampling procedures and equipment should be designed and audited and how the uncertainty of the sampling can be estimated Pierre Gy has developed a comprehensive theory that treats sampling for chemical analysis This short course is based on his work Target Audience Analytical chemists designers of process analyzers chemometricians interested in the quality of data About the Instructor Pentti Minkkinen is Professor if Inorganic and Analytical Chemistry at Lappeenranta University of Technology Finland He has been teaching sampling as part of Process and Environmental Chemistry at his own University and in dedicated continuing education courses in several countries Class Schedule Introduction to Sampling will be September 22 2002 The course schedule will be as follows 7 30 8 30 Check in 8 30 10 00 Instruction 10 00 10 15 Break coffee and tea 10 15 12 00 Instruction 12 00 1 00 Lunch provided 1 00 3 00 Instruction 3 00 3 15 Break coffee tea sodas cookies 3 15 5 00 Instruction A revised schedule will be posted when exact details become available Course Fee The price schedule is shown below Early registration on or before August 15 Introduction to Sampling 1 day 295 Regular registration after August 15 Introduction to Sampling 1 day

Original URL path: http://eigenvector.com/CAC2002/Sampling.html (2016-04-27)

Open archived version from archive - CAC-2002 Intro to Sampling Registration

address Company or University Address 1 Address 2 City State or Province Zip Code or Postal Code Country Phone Number FAX Number REGISTRATION FEE Early payment before August 1 295 Regular payment after August 1 395 STUDENT DISCOUNT I m a student currently enrolled in a degree program and can provide a letter from my department chair stating this Please deduct 100 from my registration No Yes PAYMENT I ll

Original URL path: http://eigenvector.com/CAC2002/SamplingForm.html (2016-04-27)

Open archived version from archive