Typing a long list of Matlab can be tedious and frustrating.
A better way is to put the sequence of commands into a file.
If you make a file called hw.m in the same directory you are running
Matlab from, then typing "hw" to the matlab prompt causes Matlab
to do the commands in the hw.m file.  For example, if the file 
contains the lines:

n = 20;
for i=1:n
  x(i) = i;
  y(i) = i*i;
end
plot(x,y)

then typing "hw" to the Matlab prompt will produce a plot of a parabola.
Now, if you want to do it again with n=30, just edit that line and type
"hw" again.  Notice that I indented the two commands "x(i) = i;" and
"y(i) = i*i;".  These commands form the "body" of the "for loop".  Indenting
them makes it easier to see at a glance what the program is doing.  It 
is considered "good programming practice".

The command "r = randn(1,n);" produces an array with n independent standard
normal random variables.  You can access the k-th one using r(k).

The command "round" rounds a number to the nearest integer.  For example,
"round(3.4)" gives 3, round(3.6) gives 4, and round(-5.6) gives -6.

To figure out which bin R falls into, use "j = round(R/dx);".  This will
make j the bin.  Here , dx is the bin size.

The bins have j values from -k to k, including 0.  Therefore, the number
of bins is NB = 2*k+1.  You can find all the bin counts at the same time
by first creating an array to record the bin counts

NB = 2*k+1;
for bin=1:NB
   count(bin) = 0;
end
for i=1:n
  j = round(r(i)/dx);  
  bin = j + k + 1;
  count(bin) = count(bin) + 1;
end

In Matlab, all arrays start from index 1.  Therefore, we have to number
the bin counts starting from count 1, corresponding to bin -k, to count
k, corresponding to bin 0, to count 2*k+1, corresponding to bin k.  The
line "bin = j+k+1;" does this.