...
mdefine poly_model15 e*(0.97298342 + e*(-0.70321279 + e*(0.5436520 + e*(-0.27676659 + e*0.063628444))))+dummy
mdefine model15 8.0525*poly_model15(0.0)**2/kT**4/(exp(poly_hz43(0.0)/kT)-1.0)
and then defined as
model hz43model15
0.013322768 -0.0001 0.01 0.01 0.02 0.02
0.040367661 0.001 0 0 1e+20 1e+24
Again, the kT~0.13 and norm~0.04 parameters are not physical temperatures or neutron star radii. Rather, they are phenomenological fitting function parameters, but allowing them to vary can indicate how well the instrument being calibrated recovers the reference spectrum.
Using this model results in the following curve in the photon spectrum. The Beuermann, Model 15, and fitting functions are overplotted and are indistinguishable at the plotting scale.
...