Smooth a Graph with Gaussian Function

[Mansi]

Hello,

I am performing some Monte Carlo simulation, where I am simulating a decay spectrum for a specific number of nuclei.
I need to smooth every decay spectrum with a gaussian function and compare it to the experimental spectrum and calculate the variance.
And I need to perform it for all the simulated nuclei.
Every simulated decay spectrum is saved as a Graph.

Please help, on how to perform this smoothing operation on a Graph. I need to make it automated, where it can automatically generate some mean and then fit it in within those limits.

thanks
MS

---

Please read tips for efficient and successful posting and posting code

ROOT Version: 6.18
Platform: Mac OS Big Sur
Compiler: Not Provided

---

[couet]

https://root.cern/doc/master/classTGraph.html#aa978c8ee0162e661eae795f6f3a35589

[Mansi]

thanks.

Is there a way to automatically be able find the mean for the spectrum.

for example in a histogram we can do -

double meanMin, meanMax;
		meanMin = .6 * gIntensity->GetBinCenter(gIntensityP->GetMaximumBin());
		meanMax = 1.4* gIntensity->GetBinCenter(gIntensityP->GetMaximumBin());

		//printf("\nmax: %f\nmin: %f\n",meanMax,meanMin);
		
		TF1 *constrained_gaus = new TF1("constrained_gaus","gaus");
		constrained_gaus->SetParLimits(1,meanMin, meanMax);
    
		gIntensity->Fit("constrained_gaus");
		gIntensity->Draw("same");

How to do it in a Graph ?
MS

[dastudillo]

You can get the maximum (for a TGraph gr of size n) with

 Double_t max = TMath::MaxElement(n,gr->GetY());

then you can loop over all the n points of the graph and find the x corresponding to that y (there might be more than one such point, so take care of that), using GetPoint.