298 Chapter 8:Hypothesis Testing
1.0
0.8
0.6
0.4
0.2
0
Probability of accepting
H^0
n
12345 d = |σm − m^0 |
.95
FIGURE 8.2 The OC curve for the two-sided normal test for significance levelα=.05.
Asz.025=1.96, the desired probability is, from Equation 8.3.4,
(−
√
5 +1.96)− (−
√
5 −1.96)
= 1 − (
√
5 −1.96)−[ 1 − (
√
5 +1.96)]
= (4.196)− (.276)
=.392 ■
REMARK
The function 1−β(μ) is called thepower-functionof the test. Thus, for a given valueμ,
the power of the test is equal to the probability of rejection whenμis the true value. ■
The operating characteristic function is useful in determining how large the random
sample need be to meet certain specifications concerning type II errors. For instance,
supposethatwedesiretodeterminethesamplesizennecessarytoensurethattheprobability
of acceptingH 0 :μ=μ 0 when the true mean is actuallyμ 1 is approximatelyβ. That is,
we wantnto be such that
β(μ 1 )≈β
But from Equation 8.3.4, this is equivalent to
(√
n(μ 0 −μ 1 )
σ
+zα/2
)
−
(√
n(μ 0 −μ 1 )
σ
−zα/2
)
≈β (8.3.5)
Although the foregoing cannot be analytically solved forn, a solution can be obtained by
using the standard normal distribution table. In addition, an approximation forncan be
derived from Equation 8.3.5 as follows. To start, suppose thatμ 1 >μ 0. Then, because
this implies that
μ 0 −μ 1
σ/
√
n
−zα/2≤−zα/2