:1. Solution.  The estimate for the average is 10,000/100 =
100.  To get the SE of this estimate, we first need the SE of the
sum of 100 draws. This SE of this estimate is (the box SD) x 100. 
We don't know the box SD, so we use the sample SD as a
"bootstrap" estimate for it, and obtain for the SE of the sum of
the draws 20 x 10 = 200.  To get the SE for the average of the
100 draws we divide this by 100, getting 2.  The upper limit of a
95% confidence interval is two SE's above the average, or 100 +
(2X2) = 104.