for i = 1:10;
The first time through the loop, i will be assigned the value 1
Y=abs(atan((A(i-1,2)-A(i,2)/2000)));
The first time through the loop, i is 1, so the code would be trying to execute
Y=abs(atan((A(1-1,2)-A(1,2)/2000)));
which is
Y=abs(atan((A(0,2)-A(1,2)/2000)));
However, 0 is not a valid index in MATLAB.
You need to re-think how you calculate the first output.
i=i+1;
You really should have a Very Good Reason why you would alter the loop control variable within a for loop. MATLAB does define what happens, but it is seldom what most people expect would happen.
end
You are not storing the value of Y that you compute. The end result of the for loop is that Y would be the last value that was stored into Y. If you had fixed the i-1 problem, then that would correspond to the i=10 case for the for loop. The calculation for i = 10 has nothing to do with the calculation for i = 9 or i = 8 or so on, so if you are not going to store each of the values as you generate them, then you might as well not run any of those iterations and only run i=10 instead.
You should probably be considering assigning to Y(i) instead of to Y
