Re:Doubts over a ‘possible sign of life’ on Venus
by feoren on Tuesday November 03, 2020 @04:05PM
(#24983963)
They fit a TWELFTH-ORDER polynomial to the noise data. Twelfth! If you're ever fitting a twelfth-order polynomial to any data, ever, you need to seriously re-examine your underlying assumptions. I would go so far as to say that anyone fitting a 12th order polynomial to observational data is an idiot, at least when it comes to statistics.
[ Parent
| on HN ]
Re:Doubts over a ‘possible sign of life’ on Venus
by SubiculumCode on Tuesday November 03, 2020 @06:04PM
(#24985051)
Twelth order polynomials do seem extreme, but without being an expert in the context they used it, I cannot determine its appropriateness.
As an extreme, let us suppose that we know that there are 6 sources of noise spread out over the x-axis, and that the power of the noise will decrease x units away from each source. In that case, you might suspect that noise function needs a 12th order polynomial to fit the undulation pattern, just to get the necessary number of humps.