# Sample: Estimating the Mean The sampling distribution of ￣X is centered at μ, and in most applications the variance is smaller than

that of any other estimators of μ. Thus, the sample mean ￣x will be used as a point estimate for the population mean μ. Recall that σ2￣X= σ2/n, so a large sample will yield a value of ￣X that comes from a sampling distribution with a small variance. Hence, ￣x is likely to be a very accurate estimate of μ when n is large. Let us now consider the interval estimate of μ. If our sample is selected from a normal population or, failing this, if n is sufficiently large, we can establish a confidence interval for μ by considering the sampling distribution of ￣X . According to the Central Limit Theorem, we can expect the sampling distribution of ￣X to be approximately normally distributed with mean μ ￣X = μ and standard deviation σ ￣X = σ/√n. Writing zα/2 for the z-value above which we find an area of α/2 under the normal curve, we can see from Figure 9.2 that P(−zα/2 < Z < zα/2) = 1 − α, where Z =￣X− μσ/√n. Hence, P_−zα/2 <￣X− μσ/√n< zα/2_= 1− α.z1 −−zα/2 0 zα/2α/2 α /2α Figure 9.2: P(−zα/2 < Z < zα/2) = 1 − α. Multiplying each term in the inequality by σ/√n and then subtracting ￣X from each term and multiplying by −1 (reversing the sense of the inequalities), we obtain P_￣X− zα/2√σn< μ < ￣X + zα/2√σn_= 1− α. A random sample of size n is selected from a population whose variance σ2 is known, and the mean ￣x is computed to give the 100(1−α)% confidence interval below. It is important to emphasize that we have invoked the Central Limit Theorem above. As a result, it is important to note the conditions for applications that follow. Confidence Interval on μ, σ2 Known If ￣x is the mean of a random sample of size n from a population with known variance σ2, a 100(1 − α)% confidence interval for μ is given by ￣x − zα/2√σn< μ < ￣x + zα/2√σn, where zα/2 is the z-value leaving an area of α/2 to the right. For small samples selected from nonnormal populations, we cannot expect our degree of confidence to be accurate. However, for samples of size n ≥ 30, with the shape of the distributions not too skewed, sampling theory guarantees good results. Clearly, the values of the random variables ˆΘL and ˆΘU, defined in Section 9.3, are the confidence limits ˆθL = ￣x − zα/2√σnand ˆθU = ￣x + zα/2√σn. Different samples will yield different values of ￣x and therefore produce different interval estimates of the parameter μ, as shown in Figure 9.3. The dot at the center of each interval indicates the position of the point estimate ￣x for that random sample. Note that all of these intervals are of the same width, since their widths depend only on the choice of zα/2 once ￣x is determined. The larger the

for a selection of zα/2.value we choose for zα/2. the wider we make all the intervals and the more confident we can be that the particular sample selected will produce an interval that contains the unknown parameter μ. . In general. 100(1 − α)% of the intervals will cover μ.

3 793.

.

.-4.:0.013/ .3.14790 .41.0.70.

.

9 ! .:7..0 0.38001742:70 9.:3/07903472..

.

.

.

  .

 070   .

3  03.0  !* .

.

  .

3.

.

* .

 .

.

 .

.

.

.

.

.

.

 :70 ! .

.

.

.

  .

9072390306:.9-.  :9530..

.3/2:953-  70.078390803804190306:.93 17420.. 9072.3 !*  .3/9038:-97.908 04-9. 3.

.

 3  .

.

 3* .

 545:.0  8 343 .3  8 .7.3/ 9020.   7.250 41 80 3 8 800.3/42 8..943 480 .3.425:90/ 94.90/ 1742 .0 90   .

43/9438147.0 3. -04  9 8 25479.94389.341.7.250418031742.3994349090.708:9 9825479.545:.7.9144  431/03..3.40/ 90 0397... 29 %04702.  ..  .0  8.9 0 .39 94 025.0 .0 3907.43  343 1 89020..431/03.0 3907.80 9.3/428.-4.9439343 .55.

03-  . .1478.431/03.03907..

.

 3  .

.

 3  070.

.

.890 .3.3..:00.41.70.

.

:0841907.90/17423433472..3/42.7.3900844/ 708:98  0..545:.949079  4782.504190/897-:9438349944800/ 8....7.431/03.7 90.8.250841803_ 990 8.90 40.94:7 /070041.094-0.:7.-08 .3349050.9438 0.2508800.2539047:.07 1478.

.3/ .

& /0130/3\$0.7090.431/03.0298   .943   .

.

 3.3/ &  .

.

4. 3907..90  147 9.03907 41 0.7..70 41 90 8.0 41 .3/42 8.0 907 /98 /0503/ 43 43 90 .88433:70  %0/49. 41 9080 3907.20907 .9 7.3/90701470574/:.25080//1107039.:0841 .990 .0892..9 .908 90 548943 41 90 5439 0892. 3  11070398.90841905.0/1107039 3907.250  490 9.8 . 3/..20 /9  83...

.

707 90 . 43.0  8 /0907230/  %0 .

..:0 0 .4480 147 .

.

 90 3907.250800.7.3-09.90/574/:.943 41 ..79.3/ 90 2470 .9 .  90 /07 0 2.  147 .9905.20907   3 0307.. 800.33907.0.:.78.431/0390.38 90 :3343 5.8 ..439.9.0 .

.

     .

41903907.8..4.07   .