[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

RE: Tyrannosaurus was not a fat boy or girl



I also think it is important to note that curve fitting is done on and works
for the data range you have. It can often be very bad at predictions outside
the data cloud used for the curve fitting - in both directions. So the fact
that a curve seems to hit 0 at a lower age than the range they have
(estimates with small number of points anyway) means absolutely nothing.
However, the work they did seems pretty sound given their assumptions and
work. Obviously, and this is also the case, you can argue with and about
those assumptions, so please continue if you want its really worth the
discussion and is constructive and instructive. 

Predicting outside the range, especially to the right, can work well in some
cases but the objects being studied have to cooperate and it's not always
easy to know when that is or should be the case, especially when allometric
effects and biomechanical constraints - sometimes catastrophic in nature -
can creep in when you're studying things like weight, etc. Sometimes you can
keep getting bigger until you really can't and the game is over. And
sometimes organisms just change their trends big time without warning along
their life history. You often see this with sexual dimorphism, for example,
but that would be expected well within the data range (easily) for the data
being discussed.

So the curve fit in the paper is not at all problematic - at least until I
actually try it myself with their data.

R


-----Original Message-----
From: owner-DINOSAUR@usc.edu [mailto:owner-DINOSAUR@usc.edu] On Behalf Of
Paul P
Sent: Saturday, October 22, 2011 7:17 PM
To: dinosaur@usc.edu
Subject: Re: Tyrannosaurus was not a fat boy or girl

--- On Sat, 10/22/11, GSP1954@aol.com <GSP1954@aol.com> wrote:

> Also do not quite get the plot in Fig 6 in the Hutchinson
> et al. paper. It  seems to show 10 yr old Tyrannosaurus 
> not weighing anything. Odd. 

There are very few data points in the graph, so the best-fit 
lines are only crude estimates of reality. The authors are 
also estimating the age of the specimen. 

    Paul P.