[SciPy-user] histogram q's

[mcsefl]

Hi all,

Sorry to bother on the F-D bin size question.   It was easier to code than I
thought because of the scoreatpercentile function.  Even if scipy.stats
doesn't have it, here is my implementation:

from scipy.stats import
scoreatpercentile

from numpy import
sort



def
FDbinSize(X):

    """Calculates the Freedman-Diaconis bin size
for
    a data set for use in making a
histogram




Arguments:

    X:  1D Data
set




Returns:

    h:  F-D bin
size


"""

    # First Calculate the interquartile
range
    X =
sort(X)

    upperQuartile =
scoreatpercentile(X,.75)

    lowerQuartile =
scoreatpercentile(X,.25)

    IQR = upperQuartile -
lowerQuartile



    # Find the F-D bin
size

    h =
2.*IQR/len(X)**(1./3.)

    return h

-Frank
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20090306/a5c17884/attachment.html>