Posted By: GPEngine
Residuals from polyfit and LinearReg - 07/30/14 04:33
polyfit and LinearReg should have a side effect of populating a global variable, perhaps rError, representing the mean squared error of the fit.
Posted By: GPEngine
Re: Residuals from polyfit and LinearReg - 07/30/14 04:38
or some other continuous measurement of goodness of fit
Posted By: GPEngine
Re: Residuals from polyfit and LinearReg - 07/30/14 05:09
Ah, I see that polyfit Returns
Correlation coefficient, normally in the 0..1 range. Gives the similarity of the price curve and the polynom.
That'll do.