Use BAT.jl For fitting

[Shoram444]

8GOAL

Prepare fitting framework. One option is to use BAT.jl.

PROCEDURE

0nubb:

parameters to fit:

• Expected total number of signal counts $\mu^S$:

$$ \mu^S = m t \varepsilon_S S $$

$$ [S] = kg^{-1}yr^{-1} $$

$$ \mu^S = ln(2) \frac{N_A a m t }{W} \varepsilon S $$ with $$ S = \frac{1}{T_{1/2}} $$

• Expected total number of background counts $\mu^B$ :

$$ \mu^B = m t \Delta E BI $$ $$ [BI] = kg^{-1}yr^{-1}keV^{-1} $$

...

To perform the fit:

1. generate data from expected background counts which can be obtained from the background model

$\rightarrow$ for each bin $n_i = rand(Poisson(\bar{b_i}))$

2. perform BAT fit and obtain 90% CI on S
3. Convert to T12
4. repeat N times - get T12 at 90% and median

[Shoram444]

Use https://github.com/JuliaPhysics/RadiationSpectra.jl to get better fit model from histograms