[Scipy-tickets] [SciPy] #767: Add the limiting gumbel distribution to genextreme distribution when shape c=0

SciPy scipy-tickets@scipy....
Fri Oct 31 11:00:19 CDT 2008


#767: Add the limiting gumbel distribution to genextreme distribution when shape
c=0
-------------------------+--------------------------------------------------
 Reporter:  pbrod        |        Owner:  somebody
     Type:  enhancement  |       Status:  new     
 Priority:  normal       |    Milestone:          
Component:  scipy.stats  |      Version:          
 Severity:  normal       |   Resolution:          
 Keywords:               |  
-------------------------+--------------------------------------------------
Comment (by josefpktd):

 Looking at distributions.py, the comments before class
 genextreme_gen(rv_continuous) say:
 {{{
 ## Generalized Extreme Value
 ##  c=0 is just gumbel distribution.
 ##  This version does not accept c==0
 ##  Use gumbel_r for c==0
 }}}

 So this is already implemented:

 {{{
 class gumbel_r_gen(rv_continuous):
     def _pdf(self, x):
         ex = exp(-x)
         return ex*exp(-ex)
     def _cdf(self, x):
         return exp(-exp(-x))
 }}}

 Additionally there is also a left-skewed gumbel_r with:

 {{{
     def _cdf(self, x):
         return 1.0-exp(-exp(x))
 }}}

 maybe a comment could be put in the extradocs of genextrem_gen to point to
 gumbel_r for the special case.

 A quick check:
 {{{
 >>> stats.genextreme.cdf([-0.5,0,1,2],0.00001)
 array([ 0.19229604,  0.36787944,  0.6922019 ,  0.87342538])
 >>> stats.gumbel_r.cdf([-0.5,0,1,2])
 array([ 0.19229565,  0.36787944,  0.69220063,  0.87342302])
 }}}


 Josef

-- 
Ticket URL: <http://scipy.org/scipy/scipy/ticket/767#comment:1>
SciPy <http://www.scipy.org/>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list