244 Chapter 7: Parameter Estimation
Hence, a 100(1−α) percent two-sided confidence interval forμis
(
x−zα/2
σ
√
n
, x+zα/2
σ
√
n
)
wherexis the observed sample mean.
Similarly, knowing thatZ =
√
n(X−σμ)is a standard normal random variable, along
with the identities
P{Z>zα}=α
and
P{Z<−zα}=α
results in one-sided confidence intervals of any desired level of confidence. Specifically, we
obtain that
(
x−zα
σ
√
n
,∞
)
and
(
−∞,x+zα
σ
√
n
)
are, respectively, 100(1−α) percent one-sided upper and 100(1−α) percent one-sided
lower confidence intervals forμ.
EXAMPLE 7.3c Use the data of Example 7.3a to obtain a 99 percent confidence interval
estimate ofμ, along with 99 percent one-sided upper and lower intervals.
SOLUTION Sincez.005=2.58, and
2.58
α
√
n
=
5.16
3
=1.72
it follows that a 99 percent confidence interval forμis
9 ±1.72
That is, the 99 percent confidence interval estimate is (7.28, 10.72).
Also, sincez.01=2.33, a 99 percent upper confidence interval is
(9−2.33(2/3),∞)=(7.447,∞)
Similarly, a 99 percent lower confidence interval is
(−∞,9+2.33(2/3))=(−∞, 10.553) ■