TGraphErrors problem in TMinuit in pyROOT

[mu10]

Hi everyone,

I have a pyROOT code fitting a 2D plot. I want to add uncertainty along y-axis to every data point. When I add the line for errors, TMinuit stops giving me the parts “Covariance Matrix”, “Correlation Matrix” and most importantly “Minimizer is Minuit / MigradImproved” – which shows the Chi2 as well-- at the end of the output. This happens even if the fit has a successful status. Some portion of the code is below.

Cheers,
M

        opt.SetDefaultMinimizer("Minuit2"); # I don't see MigradImproved option in TMinuit2 in the manuel but it works somehow when I don't add the line marked below
        opt.SetPrintLevel(3);
        opt.SetMinimizerAlgorithm("Simplex");

...
        d = ROOT.TCanvas("d","d",800,800)
        d.SetLeftMargin(0.2) #was 0.12
        periodData = 25e-9*s_to_ns
        gr_data = ROOT.TGraphErrors(len(data))
        pedestal = 0
        for x,y in enumerate(data):
                if x == 0: pedestal = y
                gr_data.SetPoint(x,x*periodData,y)
                gr_data.SetPointError(x,0,y_err) #<------ this is the part that I am talking
        gr_data.Draw("AP")
...

        fitfunction = ROOT.TF1("myFunction",myFunction,0,1500,5)
	fitfunction.SetParName(0,'rise')
	fitfunction.SetParName(1,'width')
	fitfunction.SetParName(2,'amplitude')
	fitfunction.SetParName(3,'start')
	fitfunction.SetParName(4,'pedestal')

	fitfunction.SetParLimits(0,50,600)
	fitfunction.SetParLimits(1,100,800)
	fitfunction.SetParLimits(2,4000,100000)
	fitfunction.SetParameter(4,pedestal)

        gr_data.Fit('myFunction', 'RSEMVN') 
...

[Wile_E_Coyote]

Problems with errors in FIT

In ROOT, before you try to fit your graph or histogram, you MUST set “reasonable” initial values for ALL parameters of your function (except for some “built-in” formulas, like “gaus”, for which the standard fit procedure can automatically “guess” them), otherwise the fitting procedure may easily misbehave. So, try something like: f->SetParameters(-3000., 300.); See also the “initial pre-fit” + “final fit” trick in:

[mu10]

I actually have this for my parameters. I have updated the code above by adding those line. The code is long so I am trying to minimize it here.
Thanks

[Wile_E_Coyote]

How about:

(gr_data.Fit('myFunction', 'RSEMN')).Print('V')

[mu10]

Yes, that worked. However, it closes the detailed output which was working before. So I tried

(gr_data.Fit(‘myFunction’, ‘RSEMVN’)).Print(‘V’)

and that also worked perfectly.

Thank you!