There was also that recent fossil from Argentina (?) that appears to show at least some dinos had an avian pulmonary system.
Going from 40 to 20 tons would make a
big difference.
Henrik wrote:
Quote:
|
I think it's because when you have weight based on volume and strength based on muscle cross section it's not only easy to assume some sort of power relationship but it also makes some sort of sense.
|
Of course, but lots of bad things can happen when you extrapolate assuming the same function. I like to fit curves using untransformed data and see what the fitted function is telling me.
Ah.
Quote:
|
The problem with the traditional analysis can be traced to change in the relationship between predictor and response variables attending transformation, thereby causing measurements for large animals not to be weighted appropriately in fitting models by least squares regression. Extrapolations from statistical models obtained by back-transformation from lines fitted to logarithms are unlikely to yield reliable predictions for body size in extinct animals.
|
Without seeing the details of the actual method used, I can say that applying weighting to log-transformed data must be done very carefully. If you just log-transform your data first and run a simple linear regression, the weighting (whatever you're using: 1/x, 1/x^2) won't work right.