Hello, By default what FCN function (Is it chi-square or -log-likelihood or something else) is used by Roofit and how can I change this ? Because for some reason I get a very large value of FCN of about, 3.18337e+06.
By the way, I attached the full RooFit terminal response. Note that some lines below says,
MIGRAD FAILS TO FIND IMPROVEMENT
MACHINE ACCURACY LIMITS FURTHER IMPROVEMENT.
0]
[#1] INFO:Fitting – RooAbsOptTestStatistic::ctor(nll_CBShape1_data) fixing interpretation of coefficients of any RooAddPdf to full domain of observables
[#1] INFO:Minimization – RooAbsMinimizerFcn::setOptimizeConst: activating const optimization
** 1 **SET PRINT 1
** 2 **SET NOGRAD
PARAMETER DEFINITIONS:
NO. NAME VALUE STEP SIZE LIMITS
1 CB MEAN [GeV] 1.35000e+04 3.80000e+02 1.12000e+04 1.50000e+04
2 CB Sigma[GeV] 2.00000e+03 5.00000e+02 1.00000e+03 1.00000e+04
3 alpha1 -1.00000e+00 4.99950e-01 -1.00000e+02 -1.00000e-04
4 n1 2.00000e-01 5.00000e-02 1.00000e-01 1.00000e+01
** 3 **SET ERR 0.5
** 4 **SET PRINT 1
** 5 **SET STR 1
NOW USING STRATEGY 1: TRY TO BALANCE SPEED AGAINST RELIABILITY
** 6 **MIGRAD 2000 1
FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.
START MIGRAD MINIMIZATION. STRATEGY 1. CONVERGENCE WHEN EDM .LT. 1.00e-03
FCN=3.19105e+06 FROM MIGRAD STATUS=INITIATE 57 CALLS 58 TOTAL
EDM= unknown STRATEGY= 1 NO ERROR MATRIX
EXT PARAMETER CURRENT GUESS STEP FIRST
NO. NAME VALUE ERROR SIZE DERIVATIVE
1 CB MEAN [GeV] 1.35000e+04 3.80000e+02 0.00000e+00 6.69478e+02
2 CB Sigma[GeV] 6.77170e+03 5.00000e+02 0.00000e+00 -2.01997e+03
3 alpha1 -1.81186e+00 4.99950e-01 -6.96983e-02 6.75078e+04
4 n1 2.00000e-01 5.00000e-02 0.00000e+00 -1.82344e+04
ERR DEF= 0.5
MIGRAD FAILS TO FIND IMPROVEMENT
MACHINE ACCURACY LIMITS FURTHER IMPROVEMENT.
MIGRAD MINIMIZATION HAS CONVERGED.
FCN=3.18337e+06 FROM MIGRAD STATUS=CONVERGED 1426 CALLS 1427 TOTAL
EDM=0.0296539 STRATEGY= 1 ERROR MATRIX UNCERTAINTY 1.0 per cent
EXT PARAMETER STEP FIRST
NO. NAME VALUE ERROR SIZE DERIVATIVE
1 CB MEAN [GeV] 1.15042e+04 2.24115e+02 0.00000e+00 -1.10957e+00
2 CB Sigma[GeV] 4.47641e+03 8.55876e+01 0.00000e+00 -1.80981e+00
3 alpha1 -8.60307e-01 1.94064e-02 -0.00000e+00 -1.54579e+01
4 n1 2.26113e+00 2.80674e-02 0.00000e+00 -8.65544e-01
ERR DEF= 0.5
EXTERNAL ERROR MATRIX. NDIM= 25 NPAR= 4 ERR DEF=0.5
5.104e+04 1.507e+04 -3.832e+00 -1.283e-01
1.507e+04 7.326e+03 -1.572e+00 -1.143e+00
-3.832e+00 -1.572e+00 3.766e-04 2.667e-04
-1.283e-01 -1.143e+00 2.667e-04 7.878e-04
PARAMETER CORRELATION COEFFICIENTS
NO. GLOBAL 1 2 3 4
1 0.99939 1.000 0.780 -0.874 -0.020
2 0.99632 0.780 1.000 -0.946 -0.476
3 0.99973 -0.874 -0.946 1.000 0.490
4 0.99784 -0.020 -0.476 0.490 1.000
** 7 **SET ERR 0.5
** 8 **SET PRINT 1
** 9 **HESSE 2000
COVARIANCE MATRIX CALCULATED SUCCESSFULLY
FCN=3.18337e+06 FROM HESSE STATUS=OK 31 CALLS 1458 TOTAL
EDM=0.00047745 STRATEGY= 1 ERROR MATRIX ACCURATE
EXT PARAMETER INTERNAL INTERNAL
NO. NAME VALUE ERROR STEP SIZE VALUE
1 CB MEAN [GeV] 1.15042e+04 1.55104e+02 5.27824e-02 -2.14452e+00
2 CB Sigma[GeV] 4.47641e+03 1.25069e+02 5.37251e-03 -2.29472e-01
3 alpha1 -8.60307e-01 2.04902e-02 2.89292e-04 1.38503e+00
4 n1 2.26113e+00 2.85962e-02 3.38886e-03 -5.98506e-01
ERR DEF= 0.5
EXTERNAL ERROR MATRIX. NDIM= 25 NPAR= 4 ERR DEF=0.5
2.424e+04 1.406e+04 -2.548e+00 4.369e-01
1.406e+04 1.565e+04 -2.446e+00 -5.552e-01
-2.548e+00 -2.446e+00 4.198e-04 1.851e-04
4.369e-01 -5.552e-01 1.851e-04 8.178e-04
PARAMETER CORRELATION COEFFICIENTS
NO. GLOBAL 1 2 3 4
1 0.95889 1.000 0.722 -0.799 0.098
2 0.98826 0.722 1.000 -0.954 -0.155
3 0.99438 -0.799 -0.954 1.000 0.316
4 0.91938 0.098 -0.155 0.316 1.000
[#1] INFO:Minimization – RooAbsMinimizerFcn::setOptimizeConst: deactivating const optimization
RooRealVar::CB MEAN [GeV] = 11504.2 +/- 155.104 L(11200 - 15000)
RooRealVar::CB Sigma[GeV] = 4476.41 +/- 125.069 L(1000 - 10000)
RooRealVar::alpha1 = -0.860307 +/- 0.0204902 L(-100 - -0.0001)
RooRealVar::n1 = 2.26113 +/- 0.0285962 L(0.1 - 10)