Quote:
Originally Posted by rtomes
Quote:
|
Originally Posted by Nereid
Could you please point out the post(s) in which you answered this question? If you haven't yet answered it, please do so now.Comment: it would seem that the main Tifft paper cited (concerning "periodicities") was either not read, or misunderstood (see previous post).
If we cannot establish the consistency link between the key Tifft paper and your ATM ideas, starting with an unambiguous definition of the key term (the redshift of a galaxy), how can we even begin to evaluate (much less question or challenge) claims such as this?
|
You can collect the various periodicities reported by Tifft in the 1970s as I did. That way you check that I didn't do any selection. That I do not know everything about the relevant astronomical details is not important to a sensible statistical test. What matters is that there is no biased selection. I think that it is important to only use Tifft's data from before the time at which he developed his own theory, to avoid any possible contamination from that theory. |
Could you please clarify this?
Specifically:
* what do you consider "
Tifft's data from before the time at which he developed his own theory" to be?
* how did you conclude that Tifft's 'state model' was not a theory?
* how did you establish the validity of the reported ~72 km/s redshift periodicies conclusions (derived results) in terms of the long chain of assumptions and calculations that these derived results seem (to me) to depend upon*?
Quote:
Then you can check my calculations of the harmonics up to say a million easily if you want to. This has already been done by Pete Brown in Australia (who also pointed out that my harmonics calculation is the same as a known series) and so I am sure that they are right.
Then you can use the formula (1+z)^h=2 to solve for z from each of the strong harmonics that are calculated. The result ought to be the graphic that I posted with the various z values (and some zc values for comparison to redshifts quoted in km/s).
You can then do a statistical check on the matching of the two sets in the range ~100 km/s to ~2 km/s which is the range in which Tifft reported periodicities.
Tifft was certainly not aware of the harmonics theory when he determined his values because it hadn't been invented for another 10 years after that.
I was not aware of any but the 72 km/s periods when I derived my formula, but anyway you can see that it is a very simple formula and cannot be engineered to fit Tifft's data.
I will describe that test in more detail because it is central to my claim. I used the list of harmonics that I posted to usenet and posted here recently. You can see from my graph that the 2.67 km/s period reported by Tifft is present in my graph but was not in that list because it was below the threshhold that I went to. I suggest sticking to that same threshhold because that list was made without awareness of Tifft's work. You may still be able to find some of these posts in usenet archives, but I do not think that matters.
If you look at the Tifft values, taking an average where he reports a similar value several times (e.g. 7.99 and 8.05 would be averaged to 8.02 km/s). Then the Tifft values can be put next to the Tomes list for obvious matches. The percentage differences should be determined in each case. I think that you will find that all are within 0.5% of my figures. However the figures range over a wide set of sizes, and it is easier to understand the percentages differences as being the best test if the logs of all the figures are taken, both mine and Tifft's. Then the figures are more or less evenly spread over an interval from log(100) to log(2) (in km/s). So we make the null hypothesis assumption that either Tomes figures are nonsense or Tiffts figures are nonsense. In case of either being nonsense there is no reason for them to agree with each other.
So the test is to be a chi-square test where the Tifft values are considered as falling within a percentage of the Tomes figures. The percentage is the one thing that is chosen to best fit the data so we say that it is a chi square test with 1 degree of freedom. If the percentage is 0.5% then the intervals are established about the Tomes figures of 0.5% which in log terms means plus or minus log(1.005) about each figure. This chosen range is therefore equal to just this proportion of the whole range: (log(100)-log(2))/log(1.005)/2/9 where the 9 is the number of figures in my list and the 2 is the two sides (+ and -) that we consider the values being within. This answer is that the defined interval is just 1/39 of the entire available space in the (log) range from 2 to 100 km/s.
In fact you find that nearly all the Tifft values fall in this tiny window. I think that it might be 6 out of 6 in the first few papers and maybe 8 out of 9 between all the 1970s papers. The formula for chi-square needs an estimate how many would fall in each region according to the null hypothesis. If there are 9 values then we would expect 9/39 in a region that is 1/39 of the Tifft values in the region within 0.5% of the Tomes values and the other 9*38/39 in the remainder. These values give expected values of 0.23 and 8.77 wheras the observed are (I think) 8 and 1.
Chi square is defined as sigma (o-e)^2/e where e=expected and o=observed. That gives a result of chi-square = 269.4 for 1 d.f.
My chi-square table has for d.f.=1, p=0.1 x2=2.706, p=.05 x2=3.841, p=.02 x2=5.412, p=.01 x2=6.635. You can see that the value 269.4 is a long way further up the table. 
|
It seems that this whole exercise rests upon an assumption, or postulate, (or similar): that the derived conclusions (about a ~72 km/s redshift periodicity) stated in Tifft's paper are valid, in some scientific sense. I understand from your posts in this thread that you are claiming that they are ... are you making such a claim?
Note that if you are making such a claim (which is most assuredly an ATM one), then I trust that you will be able to answer direct, pertinent questions on that claim.
If you are not making such a claim, then what ATM claim are you making?
*
This is the third time I am asking you this question; if you don't understand it, please ask for clarification.