Dear Bjoern,

The chi-squared values are here:

http://www.astro.ucla.edu/~wright/sne_cosmology.html ,near the bottom.

On the fourth diagram, the 2(1+z)*[sqrt(1+z) - 1]*c/H formula is a line rising from -0.04 to about 0.16 at z = 0.5 and then about level until z = 2. (using H as 65.1).

The 63.8 value is mentioned here, near the bottom too. The significance is that it could be an indication that the model used to fit it is not right, as it's quite far from the WMAP figure, but I admit that 65.1 isn't much nearer.

As you can see, the last 3 bins contain just one point each.

John Hunter.

Thanks.


As I pointed out repeatedly, merely providing a number without error margins is quite meaningless in experimental physics.


Er, that's the crucial point of binned data: that every bin contains just one point! Don't you know what "binned data" means?

__________________
The universe is big. Really big. It may seem like a long way to the corner chemist, but compared to the universe, that's peanuts. (Douglas Adams, Hitchhiker's Guide to the Galaxy)

Dear Bjoern,

The other bins contain as many as 31 data points, combined into one point, true, but the last 3 bins really have just one data point, the data is in http://www.astro.ucla.edu/~wright/sne_cosmology.html

We can argue about statistics 'till the cows come home', presumably the people in charge have SNAP have come to a similar point of view, and decided that more data is needed to reach a really definite conclusion.

John Hunter.

You are right about the bins, sorry. I didn't look at the first column of the table.

But my other point still stands: as long as we don't have error margins for the number 63.8 (or 65.1, take your pick), these numbers are quite meaningless.

__________________
The universe is big. Really big. It may seem like a long way to the corner chemist, but compared to the universe, that's peanuts. (Douglas Adams, Hitchhiker's Guide to the Galaxy)

Dear Bjoern,

OK, (Ned wright in his recently amended site, says that the different values for H, are due to different amounts of light being absorbed in the different experiments), so maybe its not as important as first thought.

The chi-squared value for d=2(1+z)sqrt[(1+z) -1] of 183.8 with 182 datapoints still means the formula can't be ruled out..now I'd like to do work on something, no doubt we'll correspond in the future!

All the best,

John Hunter.

Yes, your formula apparently indeed fits the data. Congratulations on that - I know of no other non-professional ATM proponent who has achieved something like that!

But nevertheless, my argument still stands: as long as your formulas essentially fall out of thin air and are not derived from GR, your whole model rests on quite shaky foundations.

__________________
The universe is big. Really big. It may seem like a long way to the corner chemist, but compared to the universe, that's peanuts. (Douglas Adams, Hitchhiker's Guide to the Galaxy)