Problem set on Estimation

1. It is not known what proportion p of the purchases of a certain brand of breakfast cereal
are made by women and what proportion are made by men. In a random sample of 70
purchases of this cereal, it was found that 58 were made by women and 12 were made by
men. Find the M.L.E. of p. Ans: 58/70
2. Suppose 15 rats are used in a biomedical study where the rats are injected with cancer
cells and given a cancer drug that is designed to increase their survival rate. The survival
times, in months, are 14, 17, 27,18,12,8,22,13,19 and 12. Find MLE of mean of the
population. (Assume exponential distribution) Ans: 16.2
3. Suppose that X1, . . . , Xn form a random sample from an exponential distribution for
which the value of the parameter β is unknown (β > 0). Find the M.L.E. of β.
4. Suppose that X1, . . . , Xn form a random sample from a distribution for which the p.d.f.
f(x|θ) is as follows:
f (x|θ) =𝑒 𝜃−𝑥 for x >θ,
=0 for x ≤ θ.
Also, suppose that the value of θ is unknown (−∞< θ <∞).
a. Show that the M.L.E. of θ does not exist.
b. Determine another version of the p.d.f. of this same distribution for which the M.L.E. of θ will
exist, and find this estimator.
1 1
5. X~ 𝑈 (𝜃 − 2 , 𝜃 + 2). Find MLE of 𝜃.

6. Suppose that X1, . . . , Xn form a random sample from the uniform distribution on the
interval [θ1, θ2], where both θ1 and θ2 are unknown (−∞< θ1< θ2 <∞). Find the
MOM and M.L.E.’s of θ1 and θ2.
7. Suppose that X1, . . . , Xn form a random sample from a Poisson distribution for which the
mean is unknown. Determine the M.L.E. of the standard deviation of the distribution.
8. Prove that the method of moments estimator for the parameter of a Bernoulli distribution
is the M.L.E.
9. Prove that the method of moments estimator for the parameter of an exponential
distribution is the M.L.E.
10. Prove that the method of moments estimator of the mean of a Poisson distribution is the
M.L.E.
11. Suppose that X = (X1, . . . ,Xn) form a random sample from a Poisson distribution for which
the value of the mean θ is unknown (θ > 0). Show that T =∑𝑛𝑖=1 𝑥𝑖 is a sufficient statistic for θ.
12. In each of these exercises, assume that the random variables X1, . . . , Xn form a random
sample of size n from the distribution specified in that exercise, and show that the statistic T
specified in the exercise is a sufficient statistic for the parameter.
(a) The Bernoulli distribution with parameter p, which is unknown (0<p <1); T =∑𝑛𝑖=1 𝑥𝑖
(b) The geometric distribution with parameter p, which isunknown (0<p <1); T =∑𝑛𝑖=1 𝑥𝑖
(c). The normal distribution for which the mean μ is known and the variance σ2 > 0 is unknown;
T =∑𝑛𝑖=1(𝑥𝑖 − 𝜇)2
13. Let X1,X2, ...,Xn be a random sample of size n from a population with density
1 𝑥
𝑓(𝑥, 𝜃) = 𝑒 − ⁄𝜃 , 0≤x<∞, where 𝜃 > 0 is a parameter. Are the estimators X1 and 𝑋̅ unbiased?
𝜃
Given, X1 and ̅𝑋 , which one is more efficient estimator of 𝜃?
14. Let X1,X2, ...,Xn be a random sample of size n from a distribution with density function
3
𝑓(𝑥, 𝜃) = 3𝜃𝑥 2 𝑒 −𝜃𝑥 if 0 < x < ∞
=0 otherwise.
What is the Cram´er-Rao lower bound for the variance of unbiased estimator of the parameter 𝜃?
Ans: 𝜃 2 /𝑛
15. Let X1,X2, ...,Xn be a random sample of size n from a normal population with unknown
1
mean μ and variance 𝜎 2 . Show that S2 = ∑𝑛𝑖=1(𝑥𝑖 − 𝑥̅ )2 is an unbiased estimator of 𝜎 2 . .
𝑛−1
Further, show that S2 can not attain the Cram´er-Rao lower bound.
17. Given a random sample of size 𝑛 from a population with density
2 (𝜃−𝑥)
𝑓(𝑥) = 0<𝑥<𝜃
𝜃2

=0 elsewhere
Find an estimator for 𝜃 using the method of moments.
18. Given a random sample of size 𝑛 from the two-parameter exponential distribution with
density
𝑓(𝑥) = 𝜆 𝑒 − 𝜆(𝑥− 𝛿) 𝑥>𝛿
=0 elsewhere
Find estimators for 𝜆 and 𝛿 using the method of moments.
19. On twelve days selected at random, a city’s consumption of electricity was 6.4, 4.5, 10.8, 7.2,
6.8, 4.9, 3.5, 16.3, 4.8, 7.0, 8.8 and 5.4 million kilowatt-hours. Assume that these data may be
looked upon as a random sample from a Gamma population given by
𝜆𝛼
𝑓𝑋 (𝑥) = 𝑥 𝛼−1 𝑒 −𝜆𝑥 𝑥>0
𝛤(𝛼)

Estimate the parameters α and λ using the method of moments.

20. When a broker called a random sample of eight of her clients, she got a busy signal 6.5, 10.6,
8.1, 4.1, 9.3, 11.5, 7.3 and 5.7 percent of the time. Assuming that these data may be looked upon
as a random sample from a uniform population, estimate the parameters a and b using the
method of moments.
21. Among six measurements of the boiling point of a silicon compound, the size of the error
was 0.07, 0.03, 0.14, 0.04, 0.08 and 0.03ᵒC. Assuming that these data may be looked upon as a
 random sample from the population given below, estimate the parameter 𝜃 using the method of
moments.
2(𝜃−𝑥)
𝑓(𝑥) = 0<𝑥<𝜃
𝜃2

22. Given a random sample of size 𝑛 from a Gamma population with 𝑟 = 2, obtain an estimator for
the parameter 𝛽 = 1/𝜆 by the method of maximum likelihood.
Hint: The probability density of the population is
1 −𝑥⁄𝛽
𝑓(𝑥) = 𝑥𝑒 𝑥 >0 [𝛤(2) = 1]
𝛽 2 𝛤(2)

23. On 20 very cold days, a farmer got her tractor started on the following number of tries:
{1, 3, 5, 1, 2, 1, 3, 7, 2, 4, 4, 8, 1, 3, 6, 5, 2, 1, 6, 2}
Assuming that these data may be looked upon as a random sample from the geometric
population, estimate the parameter 𝑝 using (i) the method of moments (ii) the method of
maximum likelihood.
24. Suppose that ̂ ̂2 and 𝜃
𝜃1 , 𝜃 ̂3 , are estimators of θ. We know that𝐸(𝜃
̂1 ) =E( 𝜃
̂2 )= θ and 𝐸(𝜃
̂3 ) ≠
̂1 )=12 and Var( 𝜃
θ . Var(𝜃 ̂2 ) = 10 , E(𝜃̂3 - θ)2 =6. Compare these three estimators. Which do
you prefer? Why?

25. let X1,X2, ...,Xn is a random sample from a normal distribution with mean μ and variance σ2.
Let Xmin and Xmax be the smallest and largest observations in the sample.
(a) Is (Xmin + Xmax )/2 an unbiased estimate for μ?
(b) What is the standard error of this estimate?
(c) Would this estimate be preferable to the sample mean?

26. Let three random samples of sizes n1=20, n2= 10 and n3 = 8 be taken from a population
with mean μ and variance σ2 . Let 𝑆12 , 𝑆22 and 𝑆32 be the sample variances. Show that
20𝑆12 +10𝑆22 +8 𝑆32
S2= is an unbiased estimator of σ2.
38

27. 𝑋 ̅̅̅1and 𝑆12 are the sample mean and sample variance from a population with mean μ1 and
variance σ2 1. Similarly,
̅̅̅2 and 𝑆22 are the sample mean and sample variance from a second independent population with
𝑋
mean μ2 and variance σ22 . The sample sizes are 𝑛1 and 𝑛2 , respectively.
(a) Show that 𝑋 ̅̅̅1 − 𝑋
̅̅̅2 is an unbiased estimator of μ1- μ2 .
(b) Find the standard error of ̅̅̅ ̅̅̅2 .
𝑋1 − 𝑋
 (𝑛1 −1)𝑆12 +(𝑛2 −1)𝑆22
(c) Suppose that both populations have the same variance σ2 , . Show that S2= 𝑛1 +𝑛2 −2
is an unbiased estimator of σ . 2

28. Consider the following probability density function

𝑥 −𝑥 2⁄
f(x)=𝜃 𝑒 2𝜃 , x>0; 0<θ<α
(a) It can be shown that E(x2)=2θ. Use this information to construct an unbiased estimator
for θ.
(b) Find the maximum likelihood estimator of θ and check whether it is an unbiased
estimator.

29. Let X1, . . . , Xn be a random sample from a Bernoulli trial with probability of success p.
Show that the maximum likelihood estimator is also an efficient estimator.