The function you feed in ("y" in this case) is using a vector in t that is not necessarily the same size as the "x" that is being passed in. The documentation indicates,
The function y = fun(x) should accept a vector argument x and return a vector result y, the integrand evaluated at each element of x.
but in your code if a single x were passed in, a vector of length length(t) would result, and if the length of the x that quad() chooses to pass in happens to be the same as length(t) then you are okay, but any other length(x) that quad chooses to pass in will generate an error in subtracting x-t
Also notice that you have
x=(0:0.01:t);
but you have created t as a vector. You are going to get a warning (or error) from trying to use a vector as the upper limit of a ":" operation. Did you perhaps want t(i) as the upper limit on x ?
But that x is not being used in the "y=" because x in that assignment refers to the anonymous parameter (x) for the @ function.
That leaves the vector x being passed in as the third parameter of quad, which is the "b" parameter. The "b" parameter is required to be a scalar, not a vector.
You have too many poorly defined behaviours to guess the code that would be appropriate for you.
You should be considering switching to integral2() anyhow.
