The Australian National University Research School of Finance, Actuarial Studies and Statistics

INTRODUCTORY/PRINCIPLES OF MATHEMATICAL STATISTICS (STAT2001/6039/2013/6013)

Tutorial for Week 10

1. Whenever you place an amount of money on black in the game of roulette at the casino, you double
that money with probability 18
37
19
and lose it with probability 37 .
You have decided to place $10 on black 50 times in a row.
Use the central limit theorem to approximate the probability that you will:

(a) not lose any money overall;

(b) neither gain nor lose any money overall (i.e., come out exactly even).

2. 15 resistors were randomly selected from the output of a process supposedly producing 10-ohm
resistors. The 15 resistors showed a sample mean of 9.8 ohms and a sample standard deviation of
0.5 ohms.
Find a 95% confidence interval for the true mean resistance of resistors produced by this process.
Assume that resistance measurements are normally distributed.

3. In conducting an inventory and audit of parts in a certain stockroom, it was found that, for 60 items
sampled, the audit value exceeded the book value on 45 items.
Find a 90% confidence interval for the true fraction of items in the stockroom for which the audit
value exceeds the book value.

4. Suppose that 2.6, 1.2 and 4.9 are a random sample from a normal distribution whose mean is zero
and whose variance  2 is unknown.
Derive and compute a central 99% confidence interval for  2 .

5. Let O1 and O2 be estimators of an unknown parameter  and suppose that E.O1 / D E.O2 / D ,
Var.O1 / D 12 , and Var.O2 / D 22 . Consider the estimator O3 D aO1 C .1 a/O2 .

(a) Show that O3 is an unbiased estimator of .

(b) If O1 and O2 are independent, how should the constant a be chosen in order to minimize the
variance of O3 ?
(c) How should the constant a be chosen to minimize the variance of O3 if O1 and O2 are not
independent but are such that Cov.O1 ; O2 / D c ¤ 0?