Efficient Bootstrapping and R Code for Computing M Times - 8,7,2,3,1,1000,10000

[Artusartos]

Homework Statement

I have the code:

>delta<-function(y,n,sigma,a,b,N){

+T<-rlogis(N,a,b)

+W<-sqrt(n)/(sqrt(2)*pi*sigma)*exp((-n*(y-T)^2)(S*sigma^2))

+d<-mean(T*W)/mean(T)

+return(d)

+}

And I want to compute this M times for y=8,n=7,sigma=2,a=3,b=1,N=1000,M=10000.

Homework Equations

The Attempt at a Solution

So this is what I wrote:

>my.sample<-delta(8, 7, 2,3,1,1000)

+M<-10000

+M;d<-numeric(M)

+for(i in 1:M)

+d=delta(y,n,sigma,a,b,N)

+{

>x<-sample(delta, 10000, replace=TRUE)

>M;d<-mean(x)}

But I keep getting errors, so I must be doing something wrong but I'm not sure what.

In other words, this is how it looks like in R:

> my.sample<-delta(8, 7, 2,3,1,1000)
Error in delta(8, 7, 2, 3, 1, 1000) : attempt to apply non-function
> M<-10000
> M;d<-numeric(M)
[1] 10000
> for(i in 1:M)
+ d=delta(y,n,sigma,a,b,N)
Error in rlogis(N, a, b) : object 'N' not found
> {
+ x<-sample(delta, 10000, replace=TRUE)
+ M;d<-mean(x)}
Error in x[.Internal(sample(length(x), size, replace, prob))] :
object of type 'closure' is not subsettable


I've been trying to fix this for a long time, and it's really frustrating...any help would be appreciated.

Thanks in advance

 

[mighty2000]

!

Hello,

It looks like there are a few issues with your code. First, in the line "my.sample<-delta(8, 7, 2,3,1,1000)", you are calling the function "delta" but you have not defined it yet. You need to define the function before you can use it.

Second, in the for loop, you are trying to use the variables y, n, sigma, a, b, and N, but these variables are not defined in the loop. You need to either define them within the loop or pass them as arguments to the function "delta".

Third, in the line "x<-sample(delta, 10000, replace=TRUE)", you are trying to sample from the function "delta", but you should be sampling from the output of the function. So it should be "x<-sample(my.sample, 10000, replace=TRUE)".

Finally, you are defining M twice in your code, which can cause confusion. It would be better to use a different variable name for the length of the vector, such as "n.simulations" or something similar.

I hope this helps! Let me know if you have any other questions.