Continuing the discussion from: Implementing decay time fit with event-level resolution function, confusing results

[amroo]

Continuing the discussion from Implementing decay time fit with event-level resolution function, confusing results:

According to the reply from @moneta if Method 1 is correct, then what is the interpretation of mean it can’t be of order 10^-2 given the lifetime tau is of order 10^-4 ?

[moneta]

I think the results shown in the linked past are swapped, the correct is Method1, as implemented in the linked code, and gives the result that is shown for Method 2.
This is what I am getting running the provided code now.
***Method1
from running ds_lifetime_1.C:

  RooFitResult: minimized FCN value: -2.49138e+06, estimated distance to minimum: 0.000161831
                covariance matrix quality: Full, accurate covariance matrix
                Status : MINIMIZE=0 MINOS=0 

    Floating Parameter  InitialValue    FinalValue (+HiError,-LoError)    GblCorr.
  --------------------  ------------  ----------------------------------  --------
                 delta    1.0866e+00    1.1916e+00 (+8.39e-03,-8.24e-03)  <none>
               fr_g_dt    1.3320e-03    1.3500e-02 (+1.02e-03,-9.82e-04)  <none>
                 gamma   -1.8433e+00   -2.0190e+00 (+2.30e-02,-2.36e-02)  <none>
                lambda    9.0977e-06    9.5818e-06 (+8.61e-08,-8.58e-08)  <none>
                  mean   -9.6745e-07   -9.6736e-07 (+6.91e-07,-6.92e-07)  <none>
               mean_dt    2.5730e-04    2.5883e-04 (+6.47e-06,-6.24e-06)  <none>
                   sc1    1.1801e+00    1.1808e+00 (+8.18e-03,-8.22e-03)  <none>
              sigma_dt    1.0766e-04    1.1355e-04 (+4.56e-06,-4.46e-06)  <none>
                   tau    5.0516e-04    5.0519e-04 (+1.46e-06,-1.46e-06)  <none>
                    xi    4.1211e-05    3.9689e-05 (+1.80e-07,-1.82e-07)  <none>

Method2

from running ds_lifetime_2.C:

  RooFitResult: minimized FCN value: -2.4916e+06, estimated distance to minimum: 2.38798e-05
                covariance matrix quality: Full, accurate covariance matrix
                Status : MINIMIZE=0 MINOS=70 

    Floating Parameter  InitialValue    FinalValue (+HiError,-LoError)    GblCorr.
  --------------------  ------------  ----------------------------------  --------
                 delta    1.1000e+00    1.1000e+00 (+6.42e-03,-6.30e-03)  <none>
               fr_g_dt    1.5043e-01    1.5053e-01 (+0.00e+00,-6.60e-03)  <none>
                 gamma   -2.1692e+00   -2.1694e+00 (+3.08e-02,-3.23e-02)  <none>
                lambda    7.9885e-06    7.9869e-06 (+1.10e-07,-1.13e-07)  <none>
                  mean   -2.6160e-02   -2.7021e-02 (+1.00e-02,-1.00e-02)  <none>
               mean_dt    6.1841e-05    6.1840e-05 (+3.11e-07,-3.05e-07)  <none>
                   sc1    1.1747e+00    1.1743e+00 (+8.33e-03,-8.33e-03)  <none>
              sigma_dt    1.0690e-05    1.0693e-05 (+1.87e-07,-1.95e-07)  <none>
                   tau    5.0612e-04    5.0617e-04 (+1.49e-06,-1.48e-06)  <none>
                    xi    3.8489e-05    3.8489e-05 (+2.00e-07,-2.03e-07)  <none>

Lorenzo

[amroo]

@moneta
Thanks for such a swift response.
Actually, It was my mistake. Results are not swapped, but the script names are:
ds_lifetime1.C is for Method 2 and
ds_lifetime2.C is for Method 1.

Uploading by name correction:
method1.C (9.1 KB)
method2.C (9.1 KB)