In: Statistics and Probability
4. Many realistic applications involve sampling without replacement. For example, in manufacturing, quality control inspectors sample items from a finite production run without replacement. For such a finite population, we have to
adjust the value of σ(X). Take (without replacement) samples of size 2 from the above population of N= 5 objects {1, 2, 3, 4, 5}
(1,2) (1,3) (1,4) (1,5) (2,3), (2,4), (2,5)................ ______________________________________________________
a. How many such samples are possible?
b. List all the X ’s, i.e. the means of all these samples of size 2 _______________________________________________________
c. Find the mean of these means, i.e. E(X) d. Is E(X)=μ?
e. Find the standard errorσ(X), i.e. the std. deviation of all these means
2
f. Show that the standard error equals
Note: N − n is called the finite population correction factor. Typically it is N −1
used when the sample size n is greater than 5% of the finite population size.