Gamfit function not working

Hi, can anyone tell what am I doing wrong in this or give the correct code to find confidence interval for Gamma function?

Here's the code that I wrote

a = 2; b = 4;

r = gamrnd(a,b,100,1);

[p,ci] = gamfit(r);

Answer 1

[Star Strider]

That code runs for me without error.

Is there a problem?

EDIT —

The confidence intervals are for the parameters.

Answer 2

[John D'Errico]

Do you seriously expect to recover the exact values of those parameters? With only 100 samples, the CI are pretty wide.

[p,ci] = gamfit(r)
p =
           2.1896399976245           3.4111990831996
ci =
          1.69026534773641          2.55050801547712
          2.83655067863624          4.56233782235144

Now, try it again, but with a larger sample size.

r = gamrnd(a,b,10000,1);
>> [p,ci] = gamfit(r)
p =
          1.97058207162021          4.11621427675147
ci =
          1.92055036089918           3.9975075019774
          2.02191714419451          4.23844607264689

It seems to be working fine. The parameters are converging to the original values. Small sample sizes are always going to be a limitation of any such parameter estimation. Remember, the gamma distribution chosen is one that is skewed.

fplot(@(x) gampdf(a,b,x))