Orthogonal Distance Regression fitting in a for loop

Share icon

Tue, 01/31/2012 - 02:34 am

Hi,

I am doing a linear fit to different columns in 2D wave. Accesing to different datasets is in for loops and because of this I also have errors in X wave, I want to do Orthogonal Distance Regression with X weightings.
When I am doing a regular line fitting I don't have any problems to append fitting results for different datasets in separate plots. But, when I choose flag /ODR=2 for ODR fitting, all the fitting results written in the resulting waves "fit_*" have zero values and there is no problems with resulting coefficients and residuals waves. Which flags I have to consider in the command line for ODR fitting?

My code looks like:

Function PlotInputDataAndFit(w2D,numberOfI,numberOfpH)
    Wave w2D                // 2-D wave containing data
 
    Variable numberOfI
    Variable numberOfpH
 
           
 //determining positions of data in columns
            Variable i, j           // Column for determining X and Y values   
            Make/O/T ionic = {"15","30","50","75","100","175"}
            String wTitle       //the title of the graph
            String graphName = ""
        For (i=0; i<numberOfI;i+=1)
            For (j=0;j<numberOfpH;j+=1)
            wTitle = "pH=" + num2str(j+3)+ ", " + "I=" + ionic[i]
            Display /W=(5,45,1053,767) w2D[][48*i+6*j+4] vs w2D[][48*i+6*j+2] as wTitle
            ModifyGraph frameStyle=frameStyle
            ModifyGraph mode=3,marker=19
           
            // /N suppresses screen updates during fitting
            // /Q suppresses history output during fitting
            //CurveFit/N/Q/NTHR=0 line w2D[][48*i+6*j+4] /X=w2D[][48*i+6*j+2] /D /R    
            CurveFit/N/Q/NTHR=0/ODR=2 line w2D[][48*i+6*j+4] /X=w2D[][48*i+6*j+2] /W=w2D[][48*i+6*j+5]/XW=w2D[][48*i+6*j+3] /D /R /I=1
            wave w=$("fit_"+ nameOfWave(w2D))
            wave wR=$("Res_"+ nameOfWave(w2D))
            //Save the coefficients
            Duplicate/O w $("fit_"+"pH" + num2str(j+3) + "I" + ionic[i]+"_"+ nameOfWave(w2D))
            Duplicate/O W_coef $("coef_"+"pH" + num2str(j+3) + "I" + ionic[i]+"_"+ nameOfWave(w2D))
            //Save residuals
            Duplicate/O wR $("Res_"+"pH" + num2str(j+3) + "I" + ionic[i]+"_"+ nameOfWave(w2D))
            //Save errors
            Duplicate/O W_sigma $("Sigma_"+"pH" + num2str(j+3) + "I" + ionic[i]+"_"+ nameOfWave(w2D))
            wave w1=$("fit_"+"pH" + num2str(j+3) + "I" + ionic[i]+"_"+ nameOfWave(w2D))
            wave wR1=$("Res_"+"pH" + num2str(j+3) + "I" + ionic[i]+"_"+ nameOfWave(w2D))
            appendToGraph w1
            removeFromGraph/z $nameOfWave(w)
           
        endfor
        endfor
End


The file with data is attached bellow and this file was imported in Igor as 2D wave.
I am running this function as e.g. PlotInputDataAndFit(wave0, 3,3).

Thanks for any suggestions.
Dejan
 DLS_Igor.txt