Array inside a function

I have the following function

U2=@(eta) i/alpha2*DV2(eta)-beta2/alpha2*W2VEC;

where W2VEC is a vector of N components and alpha2,beta2 are just numbers.

I would like to be able to evaluate U2 at N different eta, which correspond to the N components of W2VEC as well.

The way I am doing it leads to a NxN vector (every eta multiplies a N-component vector), but this is not what I want. I would like that each eta would correspond to a component of W2VEC.

Is it possible to do this having W2VEC declared already as an array of N given components? When I do U2(0), for example, I still get a vector. And what I want to obtain is a single value (corresponding to DV2(0), which is OK, and the first component of W2VEC, which fails). I know how to do this manually (using a loop and redefining U2 each time), but I would like to do it in a more sustainable way. Is that possible?

Conversely, even if W2VEC is a numeric array, is it possible to convert it to a kind of "function" so that I can access each component through the argument? (An therefore, being able to introduce this new "function" into U2)

[Adam]

You need to use component-wise operations .* and ./ rather than matrix based * and /

I'm not sure I fully understand why U2(0) still produces a vector output, but you haven't told us exactly what the sizes are of all the components and what DV2 is. Is it an array or a function?

[Jan]

@carlos g: It will be much easier to control the behavior of your function if you move it to a function, instead of the anonymous function in a one-liner.

 W2VEC is a numeric array, is it possible to convert it to a kind
 of "function" so that I can access each component through the
 argument?

This sounds like you want to use an index of the array.