Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

[Phys-l] The Pathological Quadratic Revisited.



As advertized, this was a puzzle of unusual interest.
I finally plotted root statistics for the addition of judicious amounts of Gaussian noise
of mean zero and specified SD, to each of the polynomial coefficients, and read out the
much greater contribution (about X24) that noise makes to the roots.

<http://i880.photobucket.com/albums/ac6/betwys/quadratic/24.jpg>

I show plots for the addition of noise amplitudes with mean value zero,
and SDs of 1E-5, 2E-5, 4E-5, 8E-5, 16E-5, 20E-5, and 24E-5

Brian W