Professional Documents
Culture Documents
PART VI
HYPOTHESIS TESTING
1
TESTS OF HYPOTHESIS
• A hypothesis is a statement about a
population parameter.
2
TESTS OF HYPOTHESIS
• STATISTICAL TEST: The statistical procedure
to draw an appropriate conclusion from
sample data about a population parameter.
• HYPOTHESIS: Any statement concerning an
unknown population parameter.
• Aim of a statistical test: test a hypothesis
concerning the values of one or more
population parameters.
3
NULL AND ALTERNATIVE HYPOTHESIS
• NULL HYPOTHESIS=H0
– E.g., a treatment has no effect or there is no
change compared with the previous situation.
• ALTERNATIVE HYPOTHESIS=HA
– E.g., a treatment has a significant effect or there is
development compared with the previous
situation.
4
TESTS OF HYPOTHESIS
• Sample Space, A: Set of all possible values of sample
values x1,x2,…,xn.
(x1,x2,…,xn) A
• Parameter Space, : Set of all possible values of the
parameters.
=Parameter Space of Null Hypothesis Parameter
Space of Alternative Hypothesis
= 0 1
H0:0
H1: 1
5
TESTS OF HYPOTHESIS
• Critical Region, C, is a subset of A which leads
to rejection region of H0.
Reject H0 if (x1,x2,…,xn)C
Not Reject H0 if (x1,x2,…,xn)C΄
• A test defines the critical region.
• A test is a rule which leads to a decision to fail
to reject or reject H0 on the basis of the
sample information.
6
TEST STATISTIC AND REJECTION
REGION
• TEST STATISTIC: The sample statistic on which
we base our decision to reject or not reject
the null hypothesis.
• REJECTION REGION: Range of values such
that, if the test statistic falls in that range, we
will decide to reject the null hypothesis,
otherwise, we will not reject the null
hypothesis.
7
TESTS OF HYPOTHESIS
• If the hypothesis completely specify the
distribution, then it is called a simple
hypothesis. Otherwise, it is composite
hypothesis.
• =(1, 2)
H0:1=3f(x;3, 2)
Composite Hypothesis
H1:1=5f(x;5, 2)
If 2 is known, simple hypothesis.
8
TESTS OF HYPOTHESIS
H0 is True H0 is False
Type I error
Reject H0 P(Type I error) = Correct Decision
1-
Type II error
Do not reject H0 Correct Decision
1- P(Type II error) =
11
HOW TO DERIVE AN APPROPRIATE
TEST
12
MOST POWERFUL TEST (MPT)
H0:=0 Simple Hypothesis
H1:=1 Simple Hypothesis
Reject H0 if (x1,x2,…,xn)C
The Neyman-Pearson Lemma:
L0
C x1, x2 ,, xn : k
L1
L0
Reject H0 if L k
L1 Proof: Available in
text books (e.g.
1 PL k 1
p.g.408)
13
EXAMPLES
• X~N(, 2) where 2 is known.
H0: = 0
H1: = 1
where 0 > 1.
Find the most powerful test of size .
14
Solution
n
1 1
L( | X ) (
2
) n exp{
2 2 i 1
(X i ) 2 } for 0
L( 1 | X ) 1 n
L( 0 | X )
exp{
2 i 1
2
[( X i 1 ) 2
( X i 2
0 ]}
)
n
1
exp{ i 0 1
[ 2
2 2 i 1
X ( ) ( 1
2
2
0 )]} k
2 2 ln k
2 X ( 1 0 ) sin ce 0 1
n ( 0 1 )
2 ln k ( 1 )
X 0 c
n ( 1 0 ) 2
15
Solution, cont.
• What is c?: It is a constant that satisfies
X 0 c 0
P( X c | 0 ) P( )
/ n / n
c 0 Z
n
since X~N(, 2).
For a pre-specified α, most powerful test says,
Reject Ho if
X c Z 0
n
X 0
Z
n
16
Examples
• Example2: See Bain & Engelhardt, 1992,
p.g.410 Find MPT of
Ho: p=p0 vs H1: p=p1 > p0
• Example 3: See Bain & Engelhardt, 1992,
p.g.411 Find MPT of
Ho: X~Unif(0,1) vs H1: X~Exp(1)
17
UNIFORMLY MOST POWERFUL (UMP) TEST
• If a test is most powerful against every possible value in a
composite alternative, then it will be a UMP test.
• One way of finding UMPT is to find MPT by Neyman-
Pearson Lemma for a particular alternative value, and
then show that test does not depend on the specific
alternative value.
• Example: X~N(, 2), we reject Ho if
X 0
Z
Note that this does not depend on
n
particular value of μ1, but only on the
fact that 0 > 1. So this is a UMPT of H0: = 0 vs H1: < 0.
18
UNIFORMLY MOST POWERFUL (UMP)
TEST
• To find UMPT, we can also use Monotone
Likelihood Ratio (MLR).
• If L=L(0)/L(1) depends on (x1,x2,…,xn) only
through the statistic y=u(x1,x2,…,xn) and L is an
increasing function of y for every given 0>1,
then we have a monotone likelihood ratio (MLR)
in statistic y.
• If L is a decreasing function of y for every given
0>1, then we have a monotone likelihood ratio
(MLR) in statistic −y.
19
UNIFORMLY MOST POWERFUL (UMP)
TEST
• Theorem: If a joint pdf f(x1,x2,…,xn;) has MLR
in the statistic Y, then a UMP test of size
• for H0:0 vs H1:>0 is to reject H0 if Yc
where P(Y c0)=.
• for H0:0 vs H1:<0 is to reject H0 if Yc
where P(Y c0)=.
20
EXAMPLE
• X~Exp()
H0:0
H1:>0
Find UMPT of size .
21
EXAMPLE
• Xi~Poi(), i=1,2,…,n
Determine whether (X1,…,Xn) has MLR property.
Find a UMP level α test for testing H0:=0
versus H1:<0.
22
GENERALIZED LIKELIHOOD RATIO TEST
(GLRT)
• GLRT is the generalization of MPT and
provides a desirable test in many applications
but it is not necessarily a UMP test.
23
GENERALIZED LIKELIHOOD RATIO TEST
(GLRT)
H0:0
H1: 1
r .s.
L f x1, x2 ,, xn ; f x1; , f x2 ; ,, f xn ;
24
GENERALIZED LIKELIHOOD RATIO TEST
(GLRT)
ˆ
L
0 The Generalize d Likelihood Ratio
Lˆ
GLRT: Reject H0 if 0
25
EXAMPLE
• X~N(, 2)
H0: = 0
H1: 0
Derive GLRT of size .
26
EXAMPLE
• Let X1,…,Xn be independent r.v.s, each with
shifted exponential p.d.f.:
1
f ( x | , ) exp{( x ) / }I[ , ] ( x)
where λ is known.
Find the LRT to test H0:=0 versus H1:>0.
ASYMPTOTIC DISTRIBUTION OF −2ln
• GLRT: Reject H0 if 0
• GLRT: Reject H0 if -2ln>-2ln0=c
under H 0
2 ln ~ k2
asympt.
28
TWO SAMPLE TESTS
X ~ Bin n1, p1 , r.s.
Y ~ Bin n 2 , p 2 , r.s.
H 0 : p1 p 2 p0
H1 : p1 p 2
29