5.12. Viewing the result from the LSQ fit to the decay curve¶
As a result, one obtains a table displayed by the programs internal editor (in this case no editing allowed) which has the following structure.
Result of decay curve analysis (with covariances) : Method: PLSQ
LinFit(t) = a1*X1(t) + a2*X2(t) + a3*X3(t)
i t X1(t) X2(t) X3(t) NetRate rUnc. LinFit relDev uTest
(m) (cps) (%) (cps) (%)
---------------------------------------------------------------------------
1 433.00 0.83176 0.00000 0.00000 0.0017278 23.33 | 0.0016381 5.5 0.2
2 1633.00 0.67000 0.00000 0.00000 0.0013944 28.49 | 0.0013195 5.7 0.2
3 2833.00 0.53970 0.00000 0.00000 0.0009500 40.99 | 0.0010629 -10.6 -0.3
4 4033.00 0.43474 0.00000 0.00000 0.0009361 41.57 | 0.0008562 9.3 0.2
5 5237.00 0.34994 0.00000 0.00000 0.0006306 60.85 | 0.0006892 -8.5 -0.1
6 6437.00 0.28189 0.00000 0.00000 0.0006722 57.19 | 0.0005552 21.1 0.3
7 7637.00 0.22707 0.00000 0.00000 0.0006306 60.85 | 0.0004472 41.0 0.5
8 8837.00 0.18291 0.00000 0.00000 0.0002833 133.18 | 0.0003602 -21.3 -0.2
9 10037.00 0.15710 0.00000 0.00000 0.0004382 103.13 | 0.0003094 41.6 0.3
---------------------------------------------------------------------------
LinFit: a1= 0.0019694 a2= 0.0000000 a3= 0.0000000 (given in cps !)
ra1= 16.240 ra2= 0.000 ra3= 0.000 (given in % !)
CHi2R= 8.481E-02
Prob= 0.000272 Prob= 0.000000 Prob= 0.000000 (t-test-signific. !)
The table columns are:
No. of measurement i
time duration between the Y-90/Sr-90 separation and the start time of the i-th measurement, is given in this example in m
decay factors of the defined components dependent on t; in this example, only the decay of Y-90 is considered; 0.000 indicates that this decay component has not been defined/used
\(NetRate\), net counting rates, in 1/s
relative standard uncertainty, in %
\(LinFit\), the value obtained from fitting the model for the net counting rate, in 1/s
relDev., relative deviation \((NetRate - LinFit)/LinFit \cdot 100\), in %
an u-test value, defined as \((NetRate - LinFit)/\sqrt{u^{2}(NetRate) + u^{2}(LinFit)}\). Absolute values being larger than 2, e.g. may indicate deviations from the model
After the table two rows follow showing the parameter values \(a_{i}\) and their relative standard uncertainties \({ra}_{i}\) obtained from the LSQ fitting. The first parameter, \(a_{1}\), gives the value of the Y-90 net counting rate being decay corrected to the time of the Y-90/Sr-90 separation. Chi2R is the value obtained for the reduced Chi-square.
After closing the editor window, the model itself can again be
edited under the menu item “Edit->Decay curve->model of the decay
curve” or with corresponding icon if the first result is
considered for improvement.