Rainer M. Krug
2007-Oct-24 08:13 UTC
[R] analytical solution to Sum of binominal distributed random numbers?
Frede Aakmann T?gersen wrote:> Perhaps > > http://stinet.dtic.mil/cgi-bin/GetTRDoc?AD=ADA266969&Location=U2&doc=GetTRDoc.pdf > > is something that you can use?Thanks a lot - that might help. Rainer> > > > Best regards > > Frede Aakmann T?gersen > Scientist > > > UNIVERSITY OF AARHUS > Faculty of Agricultural Sciences > Dept. of Genetics and Biotechnology > Blichers All? 20, P.O. BOX 50 > DK-8830 Tjele > > Phone: +45 8999 1900 > Direct: +45 8999 1878 > > E-mail: FredeA.Togersen at agrsci.dk > Web: http://www.agrsci.org > > This email may contain information that is confidential. > Any use or publication of this email without written permission from Faculty of Agricultural Sciences is not allowed. > If you are not the intended recipient, please notify Faculty of Agricultural Sciences immediately and delete this email. > > > > > >> -----Oprindelig meddelelse----- >> Fra: r-help-bounces at r-project.org >> [mailto:r-help-bounces at r-project.org] P? vegne af Rainer M Krug >> Sendt: 24. oktober 2007 09:11 >> Til: Charles C. Berry >> Cc: r-help >> Emne: Re: [R] analytical solution to Sum of binominal >> distributed random numbers? >> >> Hi Charles >> >> thanks for the pointing out that size and prob can be vectors >> as well - I tried it out but used 1 as the number of >> observations, assuming that and it only gave me one randon >> mumbewr (as it should be but not expected). >> >> But I was more looking at a analytical solution, as I have to >> sum up a huge number of random numbers. But I am going to try >> your solution as it should be much faster already. >> >> Thanks >> >> Rainer >> >> >> Charles C. Berry wrote: >>> ?rbinom >>> >>> only says: >>> >>> size: number of trials (zero or more). >>> >>> prob: probability of success on each trial. >>> >>> >>> But they can be vectors. >>> >>> BTW, you were aked to "PLEASE ... provide minimal, self-contained, >>> reproducible code." >>> >>> What you show cannot run without correction. >>> >>> Most likely, you intended size(n) to be the n-th element of >> the vector >>> 'size', which in R is written 'size[ n ]' . >>> >>> In which case >>> >>> sum (rbinom( length(prob) , size, prob ) ) >>> >>> works. >>> >>> Chuck >>> >>> On Tue, 23 Oct 2007, Rainer M Krug wrote: >>> >>>> Hi >>>> >>>> I have two vectors, prob and size, and I want to add the random >>>> deviates of these two, i.e. >>>> >>>> sum( >>>> sapply( >>>> 1:length(prob), >>>> function(n){ rbinom(1, size(n), prob(n) } >>>> ) >>>> ) >>>> >>>> My problem is that I have to do this for a large number of value >>>> combinations. Is there a faster way of doing this? >>>> >>>> Rainer >>>> >>>> ______________________________________________ >>>> R-help at r-project.org mailing list >>>> https://stat.ethz.ch/mailman/listinfo/r-help >>>> PLEASE do read the posting guide >>>> http://www.R-project.org/posting-guide.html >>>> and provide commented, minimal, self-contained, reproducible code. >>>> >>> Charles C. Berry (858) 534-2098 >>> Dept of >> Family/Preventive Medicine >>> E mailto:cberry at tajo.ucsd.edu UC San Diego >>> http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego >>> 92093-0901 >>> >>> ______________________________________________ >>> R-help at r-project.org mailing list >>> https://stat.ethz.ch/mailman/listinfo/r-help >>> PLEASE do read the posting guide >>> http://www.R-project.org/posting-guide.html >>> and provide commented, minimal, self-contained, reproducible code. >> ______________________________________________ >> R-help at r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >>
Matthias Kohl
2007-Oct-24 09:31 UTC
[R] analytical solution to Sum of binominal distributed random numbers?
Hello Rainer, You could also get a very exact approximation using our package distr; e.g. library(distr) B1 <- Binom(prob = 0.4, size = 20) B2 <- Binom(prob = 0.2, size = 10) B3 <- B1+B2 #convolution! plot(B3) hth, Matthias Rainer M. Krug schrieb:> Frede Aakmann T?gersen wrote: > >> Perhaps >> >> http://stinet.dtic.mil/cgi-bin/GetTRDoc?AD=ADA266969&Location=U2&doc=GetTRDoc.pdf >> >> is something that you can use? >> > > Thanks a lot - that might help. > > Rainer > > >> >> Best regards >> >> Frede Aakmann T?gersen >> Scientist >> >> >> UNIVERSITY OF AARHUS >> Faculty of Agricultural Sciences >> Dept. of Genetics and Biotechnology >> Blichers All? 20, P.O. BOX 50 >> DK-8830 Tjele >> >> Phone: +45 8999 1900 >> Direct: +45 8999 1878 >> >> E-mail: FredeA.Togersen at agrsci.dk >> Web: http://www.agrsci.org >> >> This email may contain information that is confidential. >> Any use or publication of this email without written permission from Faculty of Agricultural Sciences is not allowed. >> If you are not the intended recipient, please notify Faculty of Agricultural Sciences immediately and delete this email. >> >> >> >> >> >> >>> -----Oprindelig meddelelse----- >>> Fra: r-help-bounces at r-project.org >>> [mailto:r-help-bounces at r-project.org] P? vegne af Rainer M Krug >>> Sendt: 24. oktober 2007 09:11 >>> Til: Charles C. Berry >>> Cc: r-help >>> Emne: Re: [R] analytical solution to Sum of binominal >>> distributed random numbers? >>> >>> Hi Charles >>> >>> thanks for the pointing out that size and prob can be vectors >>> as well - I tried it out but used 1 as the number of >>> observations, assuming that and it only gave me one randon >>> mumbewr (as it should be but not expected). >>> >>> But I was more looking at a analytical solution, as I have to >>> sum up a huge number of random numbers. But I am going to try >>> your solution as it should be much faster already. >>> >>> Thanks >>> >>> Rainer >>> >>> >>> Charles C. Berry wrote: >>> >>>> ?rbinom >>>> >>>> only says: >>>> >>>> size: number of trials (zero or more). >>>> >>>> prob: probability of success on each trial. >>>> >>>> >>>> But they can be vectors. >>>> >>>> BTW, you were aked to "PLEASE ... provide minimal, self-contained, >>>> reproducible code." >>>> >>>> What you show cannot run without correction. >>>> >>>> Most likely, you intended size(n) to be the n-th element of >>>> >>> the vector >>> >>>> 'size', which in R is written 'size[ n ]' . >>>> >>>> In which case >>>> >>>> sum (rbinom( length(prob) , size, prob ) ) >>>> >>>> works. >>>> >>>> Chuck >>>> >>>> On Tue, 23 Oct 2007, Rainer M Krug wrote: >>>> >>>> >>>>> Hi >>>>> >>>>> I have two vectors, prob and size, and I want to add the random >>>>> deviates of these two, i.e. >>>>> >>>>> sum( >>>>> sapply( >>>>> 1:length(prob), >>>>> function(n){ rbinom(1, size(n), prob(n) } >>>>> ) >>>>> ) >>>>> >>>>> My problem is that I have to do this for a large number of value >>>>> combinations. Is there a faster way of doing this? >>>>> >>>>> Rainer >>>>> >>>>> ______________________________________________ >>>>> R-help at r-project.org mailing list >>>>> https://stat.ethz.ch/mailman/listinfo/r-help >>>>> PLEASE do read the posting guide >>>>> http://www.R-project.org/posting-guide.html >>>>> and provide commented, minimal, self-contained, reproducible code. >>>>> >>>>> >>>> Charles C. Berry (858) 534-2098 >>>> Dept of >>>> >>> Family/Preventive Medicine >>> >>>> E mailto:cberry at tajo.ucsd.edu UC San Diego >>>> http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego >>>> 92093-0901 >>>> >>>> ______________________________________________ >>>> R-help at r-project.org mailing list >>>> https://stat.ethz.ch/mailman/listinfo/r-help >>>> PLEASE do read the posting guide >>>> http://www.R-project.org/posting-guide.html >>>> and provide commented, minimal, self-contained, reproducible code. >>>> >>> ______________________________________________ >>> R-help at r-project.org mailing list >>> https://stat.ethz.ch/mailman/listinfo/r-help >>> PLEASE do read the posting guide >>> http://www.R-project.org/posting-guide.html >>> and provide commented, minimal, self-contained, reproducible code. >>> >>> > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >
G. Jay Kerns
2007-Oct-24 09:38 UTC
[R] analytical solution to Sum of binominal distributed random numbers?
Hi Rainer, The distr package can calculate the distribution for you: library(distr) X <- Binom(size = 7, prob = 0.3) Y <- Binom(size = 11, prob = 0.5) Z <- X + Y d(Z)( 0:18 ) # the pmf r(Z)( n = 5 ) # random variates Please note, however, that size and prob must be of length 1. Best, Jay On 10/24/07, Rainer M. Krug <rainer.r at krugs.de> wrote:> Frede Aakmann T?gersen wrote: > > Perhaps > > > > http://stinet.dtic.mil/cgi-bin/GetTRDoc?AD=ADA266969&Location=U2&doc=GetTRDoc.pdf > > > > is something that you can use? > > Thanks a lot - that might help. > > Rainer > > > > > > > > > Best regards > > > > Frede Aakmann T?gersen > > Scientist > > > > > > UNIVERSITY OF AARHUS > > Faculty of Agricultural Sciences > > Dept. of Genetics and Biotechnology > > Blichers All? 20, P.O. BOX 50 > > DK-8830 Tjele > > > > Phone: +45 8999 1900 > > Direct: +45 8999 1878 > > > > E-mail: FredeA.Togersen at agrsci.dk > > Web: http://www.agrsci.org > > > > This email may contain information that is confidential. > > Any use or publication of this email without written permission from Faculty of Agricultural Sciences is not allowed. > > If you are not the intended recipient, please notify Faculty of Agricultural Sciences immediately and delete this email. > > > > > > > > > > > >> -----Oprindelig meddelelse----- > >> Fra: r-help-bounces at r-project.org > >> [mailto:r-help-bounces at r-project.org] P? vegne af Rainer M Krug > >> Sendt: 24. oktober 2007 09:11 > >> Til: Charles C. Berry > >> Cc: r-help > >> Emne: Re: [R] analytical solution to Sum of binominal > >> distributed random numbers? > >> > >> Hi Charles > >> > >> thanks for the pointing out that size and prob can be vectors > >> as well - I tried it out but used 1 as the number of > >> observations, assuming that and it only gave me one randon > >> mumbewr (as it should be but not expected). > >> > >> But I was more looking at a analytical solution, as I have to > >> sum up a huge number of random numbers. But I am going to try > >> your solution as it should be much faster already. > >> > >> Thanks > >> > >> Rainer > >> > >> > >> Charles C. Berry wrote: > >>> ?rbinom > >>> > >>> only says: > >>> > >>> size: number of trials (zero or more). > >>> > >>> prob: probability of success on each trial. > >>> > >>> > >>> But they can be vectors. > >>> > >>> BTW, you were aked to "PLEASE ... provide minimal, self-contained, > >>> reproducible code." > >>> > >>> What you show cannot run without correction. > >>> > >>> Most likely, you intended size(n) to be the n-th element of > >> the vector > >>> 'size', which in R is written 'size[ n ]' . > >>> > >>> In which case > >>> > >>> sum (rbinom( length(prob) , size, prob ) ) > >>> > >>> works. > >>> > >>> Chuck > >>> > >>> On Tue, 23 Oct 2007, Rainer M Krug wrote: > >>> > >>>> Hi > >>>> > >>>> I have two vectors, prob and size, and I want to add the random > >>>> deviates of these two, i.e. > >>>> > >>>> sum( > >>>> sapply( > >>>> 1:length(prob), > >>>> function(n){ rbinom(1, size(n), prob(n) } > >>>> ) > >>>> ) > >>>> > >>>> My problem is that I have to do this for a large number of value > >>>> combinations. Is there a faster way of doing this? > >>>> > >>>> Rainer > >>>> > >>>> ______________________________________________ > >>>> R-help at r-project.org mailing list > >>>> https://stat.ethz.ch/mailman/listinfo/r-help > >>>> PLEASE do read the posting guide > >>>> http://www.R-project.org/posting-guide.html > >>>> and provide commented, minimal, self-contained, reproducible code. > >>>> > >>> Charles C. Berry (858) 534-2098 > >>> Dept of > >> Family/Preventive Medicine > >>> E mailto:cberry at tajo.ucsd.edu UC San Diego > >>> http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego > >>> 92093-0901 > >>> > >>> ______________________________________________ > >>> R-help at r-project.org mailing list > >>> https://stat.ethz.ch/mailman/listinfo/r-help > >>> PLEASE do read the posting guide > >>> http://www.R-project.org/posting-guide.html > >>> and provide commented, minimal, self-contained, reproducible code. > >> ______________________________________________ > >> R-help at r-project.org mailing list > >> https://stat.ethz.ch/mailman/listinfo/r-help > >> PLEASE do read the posting guide > >> http://www.R-project.org/posting-guide.html > >> and provide commented, minimal, self-contained, reproducible code. > >> > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >-- *************************************************** G. Jay Kerns, Ph.D. Assistant Professor / Statistics Coordinator Department of Mathematics & Statistics Youngstown State University Youngstown, OH 44555-0002 USA Office: 1035 Cushwa Hall Phone: (330) 941-3310 Office (voice mail) -3302 Department -3170 FAX E-mail: gkerns at ysu.edu http://www.cc.ysu.edu/~gjkerns/