Finite Difference Method to find gradient?

I need to find the gradient of an n-dimensional function to eventually optimize it bu I need to be doing it by finite difference method. I can't use the built-in matlab functions but I have no idea how to code finite difference for n-dimensions. The function should be entered as x(1) x(2) and so on (so that the loops can calculate the gradient) and the dimension of the function will be found from the size of the starting point vector. Can someone help me out in this please? I really don't know how to code this.

[Star Strider]

This Stack Exchange post should get you started: calculate Jacobian matrix without closed form or analytical form.

A second possibility: PROGRAMMING OF FINITE DIFFERENCE METHODS IN MATLAB.