[R] Integrate(dnorm) with different mean and standard deviation help

Rolf Turner rolf.turner at xtra.co.nz
Tue Jul 24 04:55:29 CEST 2012


integrate(dnorm, -1.96, 1.96, mean=2, sd=1)

Read the help for integrate!  It tells you that the
integrate function has a "..." argument which consists of
"additional arguments to be passed to f".

     cheers,

         Rolf Turner


On 24/07/12 14:23, FJ M wrote:
> I'm trying to provide different parameters to the integrate function for various probability functions. I'm using dnorm as the simplest example here. For instance integrate(dnorm, -1.96, 1.96) produces the correct answer for a normal distribution with mean 0 and standard deviation 1. I've tried two ways to use mean=2.0 and standard deviation 1, but with no luck. The examples follow.
>   
>
>> integrate(dnorm, -1.96, 1.96)
> 0.9500042 with absolute error < 1e-11
>> mean = 2.0
>> sd = 1.0
>> integrate(dnorm, -1.96, 1.96)
> 0.9500042 with absolute error < 1e-11
>> integrate(dnorm(mean=2.0,sd=1.0), -1.96, 1.96)
> Error in .Internal(dnorm(x, mean, sd, log)) : 'x' is missing
> Calls: integrate -> match.fun -> dnorm
> Execution halted
>   
> How do I change the built in mean=0 and standard deviation=1 for dnorm using integrate?



More information about the R-help mailing list