CHAPTER 6

Section 6.1

1.
a. We use the sample mean, to estimate the population mean .

b. We use the sample median, (the middle observation when arranged in ascending
order).

c. We use the sample standard deviation,

d. With “success” = observation greater than 10, x = # of successes = 4, and

e. We use the sample (std dev)/(mean), or

2.

a. ;

b.

The estimate would be .

c.

221
 Chapter 6: Point Estimation

3.
a. We use the sample mean,

b. Because we assume normality, the mean = median, so we also use the sample mean
. We could also easily use the sample median.

c. We use the 90th percentile of the sample:

.

d. Since we can assume normality,

e. The estimated standard error of

4.
a. With X = # of T’s in the sample, the estimator is so .

b. Here, X = # in sample without TI graphing calculator, and x = 16, so

5. Let θ = the total audited value. Three potential estimators of θ are , ,

and . From the data, = 374.6, = 340.6, and = 34.0. Knowing N = 5,000 and T

= 1,761,300, the three corresponding estimates are ,

, and

6.
a. Let for i = 1, .., 31. It is easily verified that the sample mean and sample sd
b.
c. of the are and Using the sample mean and sample sd to
estimate and , respectively, gives and (whence
).

222
 Chapter 6: Point Estimation

d. . It is natural to estimate E(X) by using and in place of

and in this expression:

7.

a.

b.

c. 8 of 10 houses in the sample used at least 100 therms (the “successes”), so

d. The ordered sample values are 89, 99, 103, 109, 118, 122, 125, 138, 147, 156, from

which the two middle values are 118 and 122, so

8.
a. With p denoting the true proportion of non-defective components,

b. P(system works) = p2 , so an estimate of this probability is

223
 Chapter 6: Point Estimation

9.
a. so is an unbiased estimator for the Poisson parameter ;
since n = 150,

b. , so the estimated standard error is

10.

a. , so and

; since

b. , so .

11.

a. .

b.

and the standard error is the square root

of this quantity.

c. With , , , , the estimated standard error is

d.

e.

224
 Chapter 6: Point Estimation

12.

13.

14.
a.

b. .

Setting the derivative with respect to equal to 0 yields , from

which .

15.

a. implies that . Consider . Then

, implying that is an unbiased

estimator for .

b. , so

225
 Chapter 6: Point Estimation

16.
a. min(xi) = 202 and max(xi) = 525, so the estimate of the number of planes manufactured is
max(xi) - min(xi) + 1 = 525 – 202 + 1 = 324.

b. The estimate will equal the true number of planes manufactured iff min(x i) =  and
max(xi) = , i.e., iff the smallest serial number in the population and the largest serial
number in the population both appear in the sample. The estimator is not unbiased. This
is because max(xi) never overestimates  and will usually underestimate it (unless
max(xi) = ) , so that E[max(xi)] < . Similarly, E[min(xi)] >  ,so E[max(xi) - min(xi)] <
 -  + 1; The estimate will usually be smaller than  -  + 1, and can never exceed it.

17.

a.

b. For the given sequence, x = 5, so

18.

a. , so the bias of the estimator is ; thus

tends to overestimate .

b. , so with ,

.
c.

226
 Chapter 6: Point Estimation

19.

a. , so and the

estimate is .

b. , as desired.

c. Here so and .

Section 6.2

20.

a. We wish to take the derivative of , set it equal to zero and solve

for p. ; setting this equal to zero

and solving for p yields . For n = 20 and x = 3,

b. ; thus is an unbiased estimator of p.

c.

227
 Chapter 6: Point Estimation

21.

a. and , so the

moment estimators and are the solution to ,

. Thus , so once has been determined

is evaluated and then computed. Since ,

, so this equation must be solved to obtain .

b. From a, , so , and

from the hint, . Then .

22.
a. , so the moment estimator is the

solution to , yielding . Since

b. , so the log likelihood is

. Taking and equating to 0 yields ,

so . Taking for each given yields ultimately .

23. For a single sample from a Poisson distribution,

, so

. Thus

228
 Chapter 6: Point Estimation

. For our problem,

is a product of the x sample likelihood and the y sample

likelihood, implying that , and (by the invariance principle) .

24.

a. . The

natural log of the likelihood function is . Taking the

derivative wrt and equating to 0 gives , so and .

The mle is therefore , which is identical to the unbiased estimator suggested in

Exercise 15.

b. For x > 0 the cdf of X if is equal to . Equating this

to .5 and solving for x gives the median in terms of : implies that

, so . The mle of is therefore

25.
a. , so and

(this is not s).

b. The 95th percentile is , so the mle of this is (by the invariance principle)
.

26. The mle of is (by the invariance principle)

229
 Chapter 6: Point Estimation

27.

a. , so the log likelihood is

. Equating both and to 0

yields and , a very difficult

system of equations to solve.

b. From the second equation in a, , so the mle of is .

28. We wish to take the derivative of with respect to p, set it equal to

zero, and solve for p: . Setting

this equal to zero and solving for p yields . This is the number of successes over
the total number of trials, which is the same estimator for the binomial in exercise 6.20. The
unbiased estimator from exercise 6.17 is , which is not the same as the
maximum likelihood estimator.

230
 Chapter 6: Point Estimation

29.
a. The joint pdf (likelihood function) is

Notice that iff ,

and that .

Thus likelihood =

Consider maximization wrt . Because the exponent is positive, increasing will

increase the likelihood provided that ; if we make larger than ,
the likelihood drops to 0. This implies that the mle of is . The log
likelihood is now . Equating the derivative wrt to 0 and solving

yields .

b. and , so

30. The likelihood is where

. We know , so by the invariance

principle for n = 20, y = 15.

Supplementary Exercises

31. Substitute k = ε/σY into Chebyshev’s inequality to write P(|Y – μY| ≥ ε) ≤ 1/(ε/σY)2 = V(Y)/ε2.

Since and , we may then write . As n →

∞, this fraction converges to 0, hence → 0, as desired.

231
 Chapter 6: Point Estimation

32.

a. for

, so .

b. While is not unbiased, is, since

33.
23.5 26.3 28.0 28.2 29.4 29.5 30.6 31.6 33.9 49.3
25.7 25.8 26.4 27.0 27.5
23.5 23.5 24.9 26.5 28.7 36.4
5 5 5 5 5
27.1 27.2 27.8 28.4 28.9
26.3 26.3 27.9 30.1 37.8
5 5 5 5 5
30.9 38.6
28.0 28.0 28.1 28.7 28.75 29.3 29.8
5 5
31.0 38.7
28.2 28.2 28.8 28.85 29.4 29.9
5 5
30.6 39.3
29.4 29.4 29.45 30.0 30.5
5 5
30.0 30.5
29.5 29.5 31.7 39.4
5 5
32.2 39.9
30.6 30.6 31.1
5 5
32.7 40.4
31.6 31.6
5 5
33.9 33.9 41.6
49.3 49.3

There are 55 averages, so the median is the 28th in order of increasing magnitude. Therefore,

232
 Chapter 6: Point Estimation

34.
, and

. To find the minimizing value of K, take

and equate to 0; the result is ; thus the estimator which minimizes MSE is neither

the unbiased estimator (K = 1) nor the mle .

35. Let x1 = the time until the first birth, x2 = the elapsed time between the first and second births,
and so on. Then .

Thus the log likelihood is . Taking and equating to 0 yields

. For the given sample, n = 6, x1 = 25.2, x2 = 41.7 – 25.2 = 16.5, x3 = 9.5, x4 =

4.3, x5 = 4.0, x6 = 2.3; so

and .

233
 Chapter 6: Point Estimation

36.

a. The likelihood is .

The log likelihood is thus . Taking and

equating to zero gives . Substituting these estimates of the into the log

likelihood gives

. Now taking , equating to zero, and

solving for gives the desired result.

, but

. Thus

, so the mle is definitely not unbiased; the expected value of

the estimator is only half the value of what is being estimated!

37. Let . Then E(cS) = cE(S), and c cancels with the two factors and the

square root in E(S), leaving just . When n = 20, . and

, but . Straightforward
calculation gives c = 1.0132.

234
 Chapter 6: Point Estimation

38. With and , . The

are, in increasing order, .02, .02, .08, .22, .32, .42, .53, .54, .65, .81, .91, 1.15,
1.17, 1.30, 1.54, 1.54, 1.71, 2.35, 2.92, 3.50. The median of these values is

. The estimate based on the resistant estimator is then .

This estimate is in reasonably close agreement with s.

235