[R] R's integrate function

[Charles C. Berry]

On Tue, 30 Sep 2008, Susanne Pfeifer wrote:

> Hi,
>
> sorry, I was a bit confused and my function looked slightly different
> from the one I posted before:


In any case, you will not get far trying to integrate this one for values 
of z like 12000 without doing some homework. Simply plugging the function 
into 'integrate' won't get an accurate answer.

The function you have has a narrow spike with a lot of mass near the 
origin when 
z=12000.

The answer you got is off by orders of magnitude. (Try changing the limits 
of integration to (0,0.1) and compare to what you got with (0,1) to see 
this).

You'll need to come up with a better method of finding this integral.

HTH,

Chuck


>
> Susanne Pfeifer schrieb:
>>>> integrate(function(y,z){
>>> +   sapply(y, function(y,z){
>>> + integrate(function(x,z)
>>> +
>>> 1/x*dbeta(0.01,x/(0.005/1.005),(1-x)/(0.005/1.005))*dbeta(y,x/(0.005/1.005),(1-x)/(0.005/1.005))*(1-y)^z,0,1
>>>
>>> +           ,423)$value
>>> + })
>>> + },0,1,423)
>
> integrate(function(y,z){
> sapply(y, function(y,z){
> integrate(function(x,z)
> 1/x*dbeta(0.01,x/(0.01/1.01),(1-x)/(0.01/1.01))*dbeta(y,x/(0.01/1.01),(1-x)/(0.01/1.01))*(1-y)^z,0,1,12000)$value
> })
> },0,1,12000)
>
> Sorry for the confusion...
>
> ______________________________________________
> 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