Congratulations you've used reinvented the Milne model!!!!!

You can search the forums for lots of detail on the problems with Milne. The short answer is 1) it requires gravity to work differently than we think it does 2) it causes the early universe to expand rather slowly which causes problems with big bang nucleosynthesis and CMB 3) it doesn't fit with the most recent supernova measurements.

There is a variant of the Milne model called Milne-Dirac in which antimatter repels matter. In one recent thread, I was sort of defending it as "probably wrong, but not totally crazy."

Yes, twofish occasionally argues the less probable side of issues [he is old and enjoys stirring the pot now and then]. The odds favoring anti gravity from anti matter are not favorable. Gravity is almost surely mediated by an integer boson, and integer bosons appear to be indifferent to charge.

The co-moving perfect cosmological principle is also consistent with the Coasting Cosmology model with equation of state:

[itex] p = -\frac{\rho c^2}{3} [/itex]

In any flat model without cosmological constant we have

[itex] \rho \propto \frac{1}{R_H^2} [/itex]

In the CC model [itex]R_H[/itex] is constant in comoving co-ordinates and therefore the energy density [itex]\rho[/itex] is constant in co-moving co-ordinates.

This makes sense to me because if you factor out the expansion of space then you would expect the energy density to remain constant as energy is conserved locally.