How to speed up this code?

I'm trying to speed up a code made of several functions and, when I run the profiler, I see that the function that takes the most time is the following in this part.

n=size(Amat,2);
for k=1:size(Amat,1)
cv(k,1)= copula_Clayton(Amat(k,1:n),alpha_0);
end

Is it possible to eliminate that for loop and make the copula_Clayton func run on every row of the matrix Amat(k,1:n) automatically?

[José-Luis]

Two things:

1. Pre-allocate cv
2. Re-write copula_Clayton to that it takes the whole Amat as input.

Please note that since we have no way of guessing at what copula_Clayton actually does, it is very difficult for us to suggest possible improvements.

Answer 1

[Azzi Abdelmalek]

You can use arrayfun, but you will notice that the for loop is faster.

cv=arrayfun(@(k) copula_Clayton(Amat(k,1:n),alpha_0),(1:size(Amat,1))','un',0);

[Giacomo]

I've already tried and I noticed it was taking longer that the loop itself.