How to use a for loop to display zeros of a function

Hello I am writing a small program in matlab to get familiar with it to find the zeroes of a function. Here is my problem, the program only displays 1 zero of the function and not all. Basically the program is displaying the last zero it finds instead of all the zeros. I am trying to modifiy it but it isn't working here is what I've got so far

    function zerosOfF=zerosOfF(f);

    N=max(size(f));
    z=f(1);
    for k=2:N
        if f(k)*f(k-1)<0; z=-z; zerosOfF=k; 
        for zerosOfF=k:N
            disp(zerosOfF);
        end
    end
    end

Answer 1

[Geoff Hayes]

You've almost have it. The problem is that the code is overwriting each index whenever a zero (of f) is found rather than saving the index to a new element in the output array.

A couple of things - z is unused and so could be removed. Also, the output array variable is named the same as the function. To avoid confusion, it should be renamed.

Try the following

 function indices=zerosOfF(f)

 N=max(size(f));

 % use this as an index into the output variable array
 at = 0;

 % initialize the output variable to an empty array
 indices = [];

 % iterate over each element
 for k=2:N

    if f(k)*f(k-1)<0; 
        at          = at+1;
        indices(at) = k; 
    end
 end

Note that you should add additional logic to handle the first and last elements as they may not be captured in the above code.

[Adam]

Hello I am having trouble with the variable "at" what is it used for? Also the empty array why is it needed?

[Geoff Hayes]

The variable at is used to insert the next zero of f into the output array indices

 at          = at+1;
 indices(at) = k;

An alternative to this would be to just append the new index to the indices array at each iteration where a zero is found (avoiding the at altogether)

 indices = [indices k];

It is a good habit to initialize variables to something before using them. In this case, I have chosen an empty array to handle the case where the input function f does not have any zeros. Suppose the code does not initialize indices to an empty array. Now try

 u=zerosOfF([1 2 3 4])

I get the following error

 Error in zerosOfF (line 3)
 N=max(size(f));

 Output argument "indices" (and maybe others) not assigned during call to
  "/Users/geoff/Development/matlab/zerosOfF.m>zerosOfF".