Indian Institute of Technology Kanpur

HSO201A Applied Probability and Statistics (2017-18-II)

PRACTICE PROBLEM SET (PPS) # 12

1. Suppose that when a signal, having value , is transmitted from location A, the value of
the signal received at another location B is normally distributed, with mean . That is,
if the value  is sent, then the value received is ( + X), where X represents noise in the
signal which is normally distributed with mean 0 and unknown, finite variance 2. To
reduce error, the same signal value is sent 9 times. If the successive values received are
5, 8.5, 12, 15, 7, 9, 7.5, 6.5 and 10.5, construct the following:

(a) A 100(1-) percent two-sided confidence interval estimate for , for  = 0.05 and 
= 0.01.
(b) A 100(1-) percent one-sided upper confidence interval estimate for , for  = 0.05
and  = 0.01.
(c) A 100(1-) percent one-sided lower confidence interval estimate for , for  = 0.05
and  = 0.01.

2. From past experience, it is known that the weights of salmon fish grown at a
commercial hatchery are normally distributed with a mean that varies from season to
season and standard deviation that remains fixed at 0.3 pounds. If we want to be 95
percent certain that our estimate of the present season’s mean weight of salmon fish is
correct to within  0.1 pounds, how large a sample of salmon fish is needed?

3. Two different types of electrical cable insulation have recently been tested to determine
the voltage level at which failures tend to occur. When cable specimens were subjected
to an increased voltage stress in a laboratory experiment, failures for the two types of
cable insulation occurred at the following voltages:

Type A Type B
36 54 52 60
44 52 64 44
41 37 38 48
53 51 68 46
38 44 66 70
36 35 52 62
34 44

Suppose that it is known that the amount of voltage that cables of type A insulation can
withstand is normally distributed, with unknown mean A and known variance A2 =
40, whereas the corresponding distribution for type B insulation is normal, with
unknown mean B and known variance B2 = 100.

(a) Determine a 95 percent two-sided confidence interval estimate for (A – B). (Hint:
Note that: (𝑋A – 𝑋B)  N(A – B, + ), where 𝑋A and 𝑋B are the sample means,

1
 and n and m are the sample sizes, of the random samples for cables of type A and type
B insulation respectively.)

(b) Determine a value that we can assert, with 95 percent confidence, exceeds (A – B).

4. Consider Problem 3 above. Suppose that the variances in the amount of voltage that
cables of type A and type B insulations can withstand are equal and unknown (A2 =
B2 = 2 < ∞). Determine a 90% two-sided confidence interval for the difference in
means (A – B). Also, determine a 95 percent one-sided upper confidence interval for
( ) ( ) 2 2
(A – B). (Hint: Note that: +  𝜒( ) , where SA and SB are the
sample variances of the random samples for cables of type A and type B insulation
respectively.)

5. A machine produces standardized bolts, with very small variance in their thickness.
Suppose that 10 bolts were chosen at random and their thickness was measured, as below,
in inches:

0.123 0.133
0.124 0.125
0.126 0.128
0.120 0.124
0.130 0.126

Find 90% confidence intervals (two-sided, one-sided upper and one-sided lower) for the
( )
variance of the thickness of a bolt produced by the machine. (Hint: Note that: 
2
𝜒( ) , where S is the sample variance.)

6. The following are the number of steps walked in each day of the last week (i.e. last seven
days): 6822 5333 7420 7432 6252 7005 6752

Suppose that the number of steps walked daily are independent and normally distributed
random variables, each with mean  and variance 2 (< ∞). Provide a prediction interval
which, with 95% confidence, will contain the number of steps that will be walked
tomorrow. (Hint: We need to predict Xn+1, given the random sample X1, X2,…, Xn. Now,
note that given our assumptions, Xn+1  N(, 2), 𝑋  N(, 2/n), and Xn+1 and 𝑋 are
( )
independent. Now, use the probability distributions of (Xn+1 – 𝑋 ) and to
determine the prediction interval for Xn+1, where S2 is the sample variance.)