Skip to content

C.D.F

3 messages · pari hesabi, Matthias Kohl, Rolf Turner

#
Dear Diba,

you could try package distr; eg.

library(distr)
G1 <- Gammad(scale = 0.7, shape = 0.5)
G2 <- Gammad(scale = 2.1, shape = 1.7)
G3 <- G1+G2 # convolution
G3

For the convolution exact formulas are applied if available, otherwise 
we use FFT; see also http://www.jstatsoft.org/v59/i04/ (will appear 
soon) resp. a previous version at http://arxiv.org/abs/1006.0764

hth
Matthias

Am 11.08.2014 um 10:17 schrieb pari hesabi:

  
    
#
On 11/08/14 20:17, pari hesabi wrote:
Is this homework?  The list has a no homework policy.

cheers,

Rolf Turner