[SciPy-User] inverse function of a spline

josef.pktd@gmai... josef.pktd@gmai...
Sat Oct 1 22:36:52 CDT 2011


An example of a possible usecase

background
I wrote a function for Lilliefors test of normality (Kolmogorov Smirnov test
for normal distribution based on estimated mean and variance). (The reported
power of the test is not great.)
The p-values are non-standard and tabulated, by sample size and by
probability (only 0.001 to 0.2)

To safe work later on for other tests, I wrote a simple class, but I
restrict to linear interpolation.

Using two interp1d or using Rbf works fine, using interp2d gives me a
warning - cannot add more knots - and the interpolated values are not
correct.

Since I never played much with interp2d, or interpolate.bisplrep or
interpolate.BivariateSpline, human error is possible.

Josef
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/scipy-user/attachments/20111001/94eaf1f2/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tabledist.py
Type: text/x-python
Size: 5107 bytes
Desc: not available
Url : http://mail.scipy.org/pipermail/scipy-user/attachments/20111001/94eaf1f2/attachment.py 


More information about the SciPy-User mailing list