MathGroup Archive 1998

[Date Index] [Thread Index] [Author Index]

Search the Archive

CDF[MultinormalDistribution[...]] only for n < 3?


Hi	Has anyone got a fix for the (apparent) bug in the standard stats
package that keeps the CDF of the multivariate normal distribution from
being computed in 3 and more dimensions?  I find the following:In[1]:=
<<Statistics`MultinormalDistribution`In[2]:=
CDF[MultinormalDistribution[{0,0},IdentityMatrix[2]],{0,0}]Out[2]=
0.25In[3]:=
CDF[MultinormalDistribution[{0,0,0},IdentityMatrix[3]],{0,0,0}]Solve::svars:
Equations may not give solutions for all "solve" variables.Out[3]=
CDF[MultinormalDistribution[{0, 0, 0}, {{1, 0, 0}, {0, 1, 0}, {0, 0,
1}}], {0, 0, 0}]	Best Regards,		Brian

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own



  • Prev by Date: beep after calculation
  • Next by Date: Re: Corrupted NB
  • Previous by thread: Re: beep after calculation
  • Next by thread: Can I get ComplexExpand to really work?