Linear fit returns parameters equal to zero

Newbie
1

I’m trying to perform a simple linear fit but the parameters I get are both equal to zero. I previously used this code for another set of data and it worked, so I really don’t understand what I’m doing wrong :frowning:

#include "TCanvas.h"
#include "TFile.h"
#include "TH1.h"
#include "TF1.h"
#include "TAxis.h"
#include "TStyle.h"
#include <cmath>
#include <string>
#include <iostream>
using namespace std;

void calibrazione() {

    gStyle->SetOptFit(0011);
    TCanvas *canvas=new TCanvas("canvas", "canvas", 800, 800);

    
    string names[4] = {"Na 1", "Na 2", "Co 1", "Co 2"}; // Names for each data point



    double x[4] = {2798.3,6732.3,6179,6988};
    double y[4] = {511,1274, 1173.2,1332.5}; 
    double ex[4]={0.3,0.8,1,1};
    double ey[4]={0};

    TGraphErrors *gr = new TGraphErrors(4,x,y,ex,ey);
    gr->SetMarkerStyle(21);
    gr->SetMarkerSize(1); 
    gr->Draw("ape");
    TF1 *fitFunc = new TF1("fitFunc", "pol1");
    gr->Fit("fitFunc");

    gr->SetTitle("Calibration; Position; Energy [keV]");
     // Add labels for each data point
     for (int i = 0; i < 4; i++) {
        double xpos1 = x[i];
        double ypos1 = y[i]+50;
        TString label1 = names[i].c_str(); 
        TLatex *latex1 = new TLatex(xpos1, ypos1, label1);
        latex1->SetTextSize(0.02);
        latex1->SetTextAlign(22); 
        latex1->Draw();
    }
  
    canvas->Update();
    
    TCanvas *canvas1=new TCanvas("canvas1", "canvas", 800, 800);
    TGraphErrors *residual= new TGraphErrors;

   
    for (int i=0; i<4; i++){
        residual->SetPoint(i , x[i], y[i]-fitFunc->Eval(x[i]));
        Double_t eyr = gr->GetErrorY(i);
        residual->SetPointError(i, 0, eyr);

    }
    residual->SetTitle("Residui (ch0)");
    residual->SetMarkerStyle(21);
    residual->Draw("ape");

    // Determine the range of the residual plot
    double xmin = residual->GetXaxis()->GetXmin();
    double xmax = residual->GetXaxis()->GetXmax();

    // Draw a dotted horizontal line at y=0 within the range of the residual plot
    TLine *line = new TLine(xmin, 0, xmax, 0);
    line->SetLineStyle(2); // Dotted line style
    line->Draw();

     // Add labels for each data point
    for (int i = 0; i < 4; i++) {
        double xpos = x[i]+0.0006;
        double ypos = y[i] - fitFunc->Eval(x[i]); 
        TString label = names[i].c_str(); 
        TLatex *latex = new TLatex(xpos, ypos, label);
        latex->SetTextSize(0.02);
        latex->SetTextAlign(22); 
        latex->Draw();
    }


    return;
}
2

It is because you errors on Y (ey) are 0.
try double ey[4]={.1,.1,.1,.1};
or use TGraph instead of TGraghErrors
Also note your points are not ordered.

3

Thank you! Since the errors were actually equal to zero I set them to a very small value and now it works :slight_smile:

4

And on top of that the x-errors (ex) are not zero and therefore the fits is not linear !