|
Post by theropod on Oct 18, 2013 20:35:24 GMT 5
The number "I try to establish" (again, were the F*** do you take your illusions from? This is a number already established long ago, even by people you completely admire in most instances) is "data". It is the figure yielded by a published regression. Now, I don't know your definition of the term "data", but in my and everyone else's opinion it is rigorous information.
It is normal to have fluctuations, this does not mean an estimate can not be an outlier any more. Sure you understood the concept of samples at all? The bigger the sample, the less influence those fluctuations you acknowledge exist have. Ergo, the more data are used in a sample, the more reliable it gets. This is not a huge difference in this case, but it is the only difference.
|
|
|
Post by Grey on Oct 18, 2013 20:47:20 GMT 5
That's not a data from Gottfried, that's a data you got using Gottfried equation. A data that I don't refute but don't make more solid than Gottfried for the already mentionned reasons. I accept this number from these others authors educated speculations (no admiration for them, only total respect), I don't accept it as an update of Gottfried more particular extrapolation.
The bigger the sample, the better is yes, but the sample of Gottfried is already enough and precise. The fact the difference is small also explains why I don't take this as better than Gottfried's. With another comparable sample you could as well get fluctuation perhaps higher than Gottfried's.
|
|
|
Post by theropod on Oct 18, 2013 21:03:49 GMT 5
I did not use Gottfried's equation, and using an equation from a paper means following that paper's work. Didn't you even read anything of what I wrote? Gottfrieds equation somehow won't work for me, I'd appreciate some advise here.
I don't get what you mean by "fluctuations higher than Gottfrieds". Gottfrieds formula produces point estimates, which obviously differ significantly from those of a similar analysis.
You acknowledge that a bigger sample is better, and at the same time you refuse to accept a regression with a bigger sample is also better, just because the other also has a big sample. That's a 14% difference we are talking about, and yet for absolutely no reason you still all the time behave as if Gottfried's method was the "best we have".
|
|
|
Post by Grey on Oct 18, 2013 21:21:37 GMT 5
You use Gottfried equation for this bigger sample, yes or no ?
The bigger the sample, the better is yes, but the sample of Gottfried is already enough and precise. The fact the difference is small also explains why I don't take this as better than Gottfried's. With another comparable sample you could as well get fluctuation perhaps higher than Gottfried's.
Yes Gottfried has the best data to date, even though it is debattable. But you cannot outdate it.
|
|
|
Post by theropod on Oct 18, 2013 21:26:05 GMT 5
I use the sample from Casey & Pratt, 1985. I tought you knew that study?
Again, this is a regression equation of the same form as Gottfried et al., 1996. It just has a slightly bigger sample, and it is less well-known.
What makes you think that Gottfried's regression is any better? Just give me a reason, please!
|
|
|
Post by creature386 on Oct 18, 2013 21:37:29 GMT 5
Don't you both pretty much agree with 50-60 t? I don't think we need to debate that much about the methodÂ…
|
|
|
Post by Grey on Oct 18, 2013 21:42:58 GMT 5
Don't you both pretty much agree with 50-60 t? I don't think we need to debate that much about the methodÂ… That's what I told theropod earlier, but he absolutely wants to debunk Gottfried by any mean, whereas he does not get the reasons why he cannot, and why no one rigorously did it since.
|
|
Fragillimus335
Member
Sauropod fanatic, and dinosaur specialist
Posts: 573
|
Post by Fragillimus335 on Oct 18, 2013 21:54:26 GMT 5
I still back Carcharodon in this engagement.
|
|
|
Post by creature386 on Oct 18, 2013 21:57:32 GMT 5
I so much miss the first five pages of this thread, where the discussion was almost perfect.
|
|
|
Post by theropod on Oct 18, 2013 23:49:44 GMT 5
Don't you both pretty much agree with 50-60 t? I don't think we need to debate that much about the methodÂ… That's what I told theropod earlier, but he absolutely wants to debunk Gottfried by any mean, whereas he does not get the reasons why he cannot, and why no one rigorously did it since. I do not, but I suggest Gottfried is not "the best we have". Other estimates appear more consistent. How damn difficult is it to accept that maybe, I'm not some arrogant asshole who just wants to claim bias, but that instead I actually argue reasonable claims?
|
|
|
Post by Grey on Oct 18, 2013 23:57:17 GMT 5
Gottfried is the best because it is the most rigorous. Which does not undermine others estimates. Period.
Fragillimus, this is not Carcharodon anymore but Otodus or Carcharocles.
|
|
|
Post by theropod on Oct 19, 2013 1:02:17 GMT 5
Again, why do you consider Casey & Pratt (1985) any less rigorous, when their method is actually the one with the bigger sample?
|
|
|
Post by theropod on Oct 19, 2013 1:06:20 GMT 5
According to PaleoDB Otodus should have prevalence (1843 vs 1923) if they turn out to be enoughly similar. However based on tooth shape, these are quite different, Otodus obliquus teeth have cusplets, proportionally longer and narrower blades and they lack serrations.
What is the data suggesting them to be congeneric?
|
|
Fragillimus335
Member
Sauropod fanatic, and dinosaur specialist
Posts: 573
|
Post by Fragillimus335 on Oct 19, 2013 1:33:32 GMT 5
Gottfried is the best because it is the most rigorous. Which does not undermine others estimates. Period. Fragillimus, this is not Carcharodon anymore but Otodus or Carcharocles. Probably not Otodus, But yeah, it could be Carcharocles or Carcharodon. No one really knows for sure do they?
|
|
|
Post by creature386 on Oct 19, 2013 1:38:29 GMT 5
Why the "probably not Otodus". I also personally prefer Carcharocles, but why is Otodus less likely?
|
|