But putting all of that aside, I will take a narrow view of the manuscript. It proposes a distance(redshift) relation, and we can quantitatively see how well this matches the data. The proper way to do this is not by making plots, it is to compute chi^2 values from the distance moduli (mu) and covariance matrix in Union2.1:
chi^2 = (mu_observed - M - mu_theory)^T . (covariance matrix^-1) . (mu_observed - M - mu_theory)
where M is a constant that can be fit (the host-mass relation can also be fit, but failing to do so won’t affect the results much). After computing chi^2 values for LambdaCDM and HU, you can see if HU is favored or disfavored by the data compared to LambdaCDM. By my eye, HU is significantly worse, but the chi^2 values will say for sure.
Answer: In trying to give current Cosmology their best shot, I tried to use the newest Cosmological Model I could find. I was directed to use Planck-15 python package. Here is the fitting code:
from astropy.cosmology import Planck15
from astropy import constants, units
R0 = (constants.c)/(Planck15.H
d_L = (Planck15.luminosity_dist
return R0, d_L
z = np.arange(0.0,1.5,0.01)
This was an honest attempt to represent Friedmann-Lemaitre Model applied to the Supernova Survey. From my research, it implements this equation:
with six parameters (if one excludes H). By comparison, HU predicts the data without any parameters (if one excludes H, which I took from the literature as being 72).
The quality of the Friedmann-Lemaitre fitting is not relevant since the main thrust of my article is to consider that that data might be wrong (biased by the lack of an epoch-dependent G).
In any event, here is the results from the requested calculation:
The Power Divergence is 1.33 and the p-value is 1.0.
The nice but uninformative figures are here:
To my unbiased eyes..:) These predictions (not fittings) are better that the six parameters Friedmann-Lemaitre fitting. One should emphasize that HU has no parameters and FL has six!
I have to say that this is a semi-log plot and shouldn’t be compared with the distance vs z plot below.
Below are the two placed in the same plot:
I suspect the reviewer thought that HU data was intended to fit the raw data (with x). They might not had realized that I corrected the data and displayed it below.
PS - By the way, I know that it is incorrect to say that a data analysis is wrong or biased because it didn’t consider an epoch-dependent G. The reason I say that is because the theory has been censored for 12 years without a peer-review and thus not using my epoch-dependent G is a matter of choice. This is my first one and I am thankful I can reply to it here.