Presumably you know that given two points (x1,y1) and (x2,y2) you can
use Pythagoras' Theorem to work out the distance between them? So, if
you put x1 in A1 and y1 in B1, and x2 in A2 and y2 in B2, then you
could have this formula in C2 to calculate the distance:
=SQRT((A2-$A$1)^2+(B2-$B$1)^2)
Put other values in columns A and B from A3 downwards, and copy the
formula down to cover them - each value in column C is the distance
between the point co-ordinates on that row and the co-ordinates in the
first row. So, you can add these up with a SUM formula (I would
suggest that you put it in C1) to give you the total distance between
all the points and the point in row 1.
Now you could do a bit of trial and error - change the values in A1
and B1 and note the changes in C1. Make a record of them all in
columns E, F and G as you vary A1, and then as you vary B1. Perhaps
you can spot some trends over the variations? Perhaps you can draw a
graph and pull some conclusions from this?
Of course, a macro could generate different values for A1 and B1
automatically, and record the results in columns E:G for you. You
might also be able to use Solver (with ATP installed) to find the
minimum automatically. Depending on the spread of your 1000 data
points, there may be several minima, or near-minima.
Hope this gives you something to start with.
Pete
