[Scipy-tickets] [SciPy] #770: cdf, pdf formulas of general exponential are wrong

SciPy scipy-tickets@scipy....
Sun Nov 2 09:17:30 CST 2008


#770: cdf, pdf formulas of general exponential are wrong
-------------------------+--------------------------------------------------
 Reporter:  josefpktd    |        Owner:  somebody
     Type:  defect       |       Status:  new     
 Priority:  normal       |    Milestone:  0.7.0   
Component:  scipy.stats  |      Version:          
 Severity:  normal       |   Resolution:          
 Keywords:               |  
-------------------------+--------------------------------------------------
Comment (by konrad):

 The formulae in the nist gexcdf.pdf are incorrect. The equations for Ryu's
 generalised exponential distribution are given in
 http://tinyurl.com/5wk2gz
 There is a minus sign missing in the equations for both the pdf and the
 cdf.

 Incorrect equations in current version:
 {{{
 class genexpon_gen(rv_continuous):
     def _pdf(self, x, a, b, c):
         return (a+b*(1-exp(-c*x)))*exp((a-b)*x+b*(1-exp(-c*x))/c)
     def _cdf(self, x, a, b, c):
         return 1.0-exp((a-b)*x + b*(1-exp(-c*x))/c)
 }}}

 Correct equations (diff file attached):
 {{{
 class genexpon_gen(rv_continuous):
     def _pdf(self, x, a, b, c):
         return (a+b*(1-exp(-c*x)))*exp((-a-b)*x+b*(1-exp(-c*x))/c)
     def _cdf(self, x, a, b, c):
         return 1.0-exp((-a-b)*x + b*(1-exp(-c*x))/c)
 }}}

 I have also added a test to check if the pdf integrates to 1 and that the
 cdf is bounded between 0 and 1.

-- 
Ticket URL: <http://scipy.org/scipy/scipy/ticket/770#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