Tutorial 7 – MTL390
Problems
1. A toy store claims that at least 80% of girls under 8 prefer dolls over other toys. A survey of 20
randomly selected girls records T , the number who prefer dolls. We test H0 : p = 0.8, H1 : p < 0.8.
The decision rule is: reject H0 if T ≤ 12.
(a) Identify the rejection region.
(b) Define the Type I error in this context.
(c) Define the Type II error in this context.
2. Using the setting in Question 1, assume T ∼ Binomial(20, p). Compute the probability of a Type I
error for the test.
3. For the same test as Question 1 & 2, compute the probability of a Type II error when p = 0.6 and
when p = 0.4.
4. Let X be a Bernoulli random variable with P (X = x; θ) = θx (1 − θ)1−x , x = 0, 1. A sample
X1 , . . . , X10 is observed. We test H0 : θ = 41 against H1 : θ < 14 . Reject H0 if 10
P
i=1 Xi < 1.. Find
the power function of this test.
1 1
5. Let X ∼ Exp(λ). Consider testing H0 : λ = 20 against H1 : λ = 30 .. A test rejects H0 if X > 28.
Find
(a) the probability of Type I error,
(b) the probability of Type II error.
6. A sample of size 1 is taken from a population distribution P (λ). To test H0 : λ = 1 against
H1 : λ = 2, consider the nonrandomized test φ(x) = 1 if x > 3, and = 0 if x ≤ 3. Find the
probabilities of type I and type II errors and the power of the test against λ = 2. If it is required
to achieve a size equal to 0.05, how should one modify the test φ?
7. An urn contains 10 marbles, of which M are white and 10−M are black. To test that M = 5 against
the alternative hypothesis that M = 6, one draws 3 marbles from the urn without replacement. The
null hypothesis is rejected if the sample contains 2 or 3 white marbles; otherwise it is accepted. Find
the size of the test and its power.
8. Let X1 , X2 , . . . , Xn be a sample from N (µ, 1). For testing µ ≤ µ0 against µ > µ0 consider the test
function ( zα
1 if x̄ > µ0 + √ n
,
φ(x1 , x2 , . . . , xn ) = zα
0 if x̄ ≤ µ0 + √ n
.
Show that the power function of φ is a nondecreasing function of µ. What is the size of the test?
9. Let X1 , X2 , . . . , Xn be a random sample from a population with finite mean µ and finite variance
σ 2 . Suppose that µ is not known, but σ is known, and it is required to test
H0 : µ = µ0 against H1 : µ = µ1 (µ1 > µ0 ).
Let n be sufficiently large so that the central limit theorem holds, and Consider the test
(
1, x̄ > k,
φ(x1 , . . . , xn ) =
0, x̄ ≤ k,
n
1X
where x̄ = xi .
n
i=1
1
(a) Find k such that the test has (approximately) size α.
(b) What is the power of this test at µ = µ1 ?
(c) If the probabilities of type I and type II errors are fixed at α and β, respectively, find the
smallest sample size needed.
10. In Problem 9, if σ is not known, find k such that the test
(
1, x̄ > k,
φ(x1 , . . . , xn ) =
0, x̄ ≤ k,
has size α.
11. Let X1 , X2 , . . . , Xn be a random sample from N (0, σ 2 ). To test
H0 : σ = σ0 against H1 : σ ̸= σ0 ,
it is suggested that the test
( Pn Pn
1, if 2 2
i=1 xi > c1 or i=1 xi < c2 ,
φ(x1 , . . . , xn ) = Pn
0, if c2 ≤ i=1 x2i ≤ c1
be used. How will you find c1 and c2 such that the size of φ is a preassigned number α, 0 < α < 1?
What is the power function of this test?
12. Let X1 , X2 , . . . , X50 denote a random sample of size 50 from a normal distribution N (θ, 100). Find a
uniformly most powerful critical region of size α = 0.10 for testing H0 : θ = 50 against H1 : θ > 50.
13. Let X1 , X2 , . . . , X20 denote a random sample
P20of size 20 from a Poisson distribution with mean θ.
Show that the critical region C defined by i=1 Xi ≥ 4.
14. If X1 , X2 , . . . , Xn is a random sample from a beta distribution with parameters α = β = θ > 0, find
a best critical region for testing H0 : θ = 1 against H1 : θ = 2.
15. In a certain chemical process, it is very important that a particular solution to be used as a reactant
have a pH of exactly 8.20. A method for determining pH for solutions of this type is known to give
measurements that are normally distributed with mean equal to the actual pH and with a standard
deviation of 0.02. Suppose 10 independent measurements yielded the following pH values:
8.18 8.17
8.16 8.15
8.17 8.21
8.22 8.16
8.19 8.18
(a) Test the hypothesis H0 : µ = 8.20 against H1 : µ ̸= 8.20. What conclusion can be drawn at
the α = 0.10 level of significance? What about at the α = 0.05 level of significance?
(b) Suppose we wish to design a test so that if the pH were really equal to 8.20, then the probability
of not rejecting H0 is 0.95. On the other hand, if the pH differs from 8.20 by 0.03 (in either
direction), we want the probability of detecting such a difference to exceed 0.95.
(i) What test procedure should be used?
(ii) What is the required sample size?
(iii) If x̄ = 8.31, what is your conclusion?
(iv) If the actual pH is 8.32, what is the probability of concluding that the pH is not 8.20 using
the foregoing procedure?