Newton Raphson Method for beta distribution

Asked Oct 05 '15 at 05:24

Active Oct 05 '15 at 06:30

Viewed 608 times

I want to simulate 100 observations from a beta distribution using the Newton Raphson Method, where the stopping rule is |xi − xi−1| < .05 and the starting point for the algorithm is 0.5 and want to find out on average how many iterations are required until the observation is accepted

R code to simulate a sample with shape1 = 3 and shape2 = 5:

x <- rbeta(n=100, tol=1e-7, shape1=3, shape2=5)

foo<-function(x,tol=1e-7, x0=0.5,N=100)
{
  x1=x0
  h=1e-7
  i=0.5;
  p<-numeric(N)
  while(i<=N)

  { df.dx=(x(x0+h)-x(x0))/h
    x1=(x0-(x(x0)/df.dx))
    p[i]=x1
    i=i+1
    if(abs(x1 - x0) < tol) break
    x0=x1} 
  return (p[1:(i-1)])
}

When I try to call function it says function not found.

edited Oct 05 '15 at 06:30

asked Oct 05 '15 at 05:24

krithika krishnan

What is `h` in your function `f`? – Oct 05 '15 at 05:29
I want to h to be max value so it can iterate – krithika krishnan Oct 05 '15 at 05:44
1

Maybe you intend `h<-1e-7`. But what are `x1` and `x0`? What does the function return? – nicola Oct 05 '15 at 05:45
1

Please revise your function. `h`, `N`, `f`, `x0` and `x1` are not defined. – Oct 05 '15 at 05:45
It cannot work, you are calling function `f` inside itself. – Oct 05 '15 at 06:21
How else do i call it – krithika krishnan Oct 05 '15 at 06:30

Newton Raphson Method for beta distribution

0 Answers0