Unused arguments within a function in R

Question

Below is the code I have. It works for primitive functions, such as sin. However, when using a function called gllik, it returns an error in f(y0): unused argument (y0). I'm not sure how to correct this.

newton_search2 <- function(f, h, guess, conv=0.001) {
   y0 <- guess
   N = 100
   i <- 1; y1 <- y0
   p <- numeric(N)
   while (i <= N) {
      make_derivative <- function(f, h) {
       (f(y0 + h) - f(y0 - h)) / (2*h)
       }
    y1 <- (y0 - (f(y0)/make_derivative(f, h)))
    p[i] <- y1
    i <- i + 1
    if (abs(y1 - y0) < conv) break
    y0 <- y1
 }
 return (p[(i-1)])
}

The gllik function is as follows:

x <- rgamma(n=30, shape=4.5)
gllik <- function() {
 s <- sum(log(x))
 n <- length(x)
 function(a) {
   (a - 1) * s - n * lgamma(a)
   }
 }

The code I used was:

newton_search2(gllik, 0.001, mean(x), conv = 0.001)

I'm not sure how to fix the error or get the correct answer which is supposed to be 4.5 (the maximum liklihood estimate of a).

Your code is not commented and your question is not a real question. What is the code that is returning an error? — MaZe, Jul 13 '15 at 15:18
Oh, sorry. I forgot to include that. The code I put in is newton_search2(gllik, 0.001, mean(x), conv = 0.001) and then I get an error saying that y0 is an unused argument. I'm not sure how to make this work. — user114634, Jul 13 '15 at 15:20
Unless I misunderstand, gllink isn't a mathematical function. It's a function which is returning a function. It works if you use: newton_search2(gllik, 0.001, mean(x), conv = 0.001) — Mhairi McNeill, Jul 13 '15 at 15:36

Lars Lau Raket · Answer 1 · 2015-07-13T19:09:50.517

3

The problem is that gllik does not take any arguments. Furthermore, it returns a function and not a value.

Perhaps what you want to to is the following?

gllik <- function(a) {
  s <- sum(log(x))
  n <- length(x)
  return((a - 1) * s - n * lgamma(a))
}

EDIT: An alternative solution is to just use the returned function. While this type of construction is often elegant, it does seem like overkill in this case:

newton_search2(gllik(), 0.001, mean(x), conv = 0.001)

edited Jul 13 '15 at 19:09

answered Jul 13 '15 at 15:37

Lars Lau Raket

1,905
20
35

I believe that gllik should return a numeric value(s). I'm not sure how to do that. – user114634 Jul 13 '15 at 15:40
Maybe think about putting all of the variables you need in gllik. scoping allows you to not, but maybe make it a bit more explicit. or a function(a,...) – Steve Bronder Jul 13 '15 at 15:43
That worked. However, it gave me double the answer I was looking for. I think I have to take the derivative twice somehow. – user114634 Jul 13 '15 at 15:52
This seems to be a problem with your optimization method. Compare with: `optim(1, gllik, method = 'Brent', lower = 0, upper = 100, control = list(fnscale = -1))` – Lars Lau Raket Jul 13 '15 at 16:02
When I knit the R Markdown file to HTML, this code gives me different answers each time I knit it. – user114634 Jul 14 '15 at 13:09
@user114634: That is probably because you generate new random data every time? You can use set.seed(1) to generate the same random data every time. – Lars Lau Raket Jul 14 '15 at 13:20
What if I want the search to go on the first and second derivative of gllik()? Would I need to fix the while loop in my newton_search2? – user114634 Jul 14 '15 at 14:01

Unused arguments within a function in R

1 Answers1