# Stat 234; Chang.

Section 02, 391255
Ben Jacobson
March 6, 2012

1

12.1

H0 : µ = 30, 000, Ha : µ > 30, 000, σ = 7500, α = .01
7500
Critical value zc = z1α = 2.3263, x
¯c = 30000 + 2.3263 √
= 30872.3805
400
¯
β = P r(RetainH0 |µ = 31000, σ = 7500) = P r(X < 30872.3805|µ = 31000, σ = 7500)
P r(Z < 0.3403) = 0.3668.
= P r(Z < (30872.380531000))
7500

400

Thus Power = 1 − β = 0.6332

2

12.2

Here we want β=0.50 when the true mean is 31000; that is
β = P r(RetainH0 |µ = 31000, σ = 7500) =
¯ <x
P r(X
¯c |µ = 31000, σ = 7500) = .50
That means that x
¯c = 31, 000.T henzc = 31000−30000
= 43 = z1−α
7500

100

So 1 − α=0.90879 and α = 0.09121.

3

12.8

a. H0 : π = 13
b. Ha : π 6= 31
c. P value using continuity correction:  

28 − .5 − 24
7
= 2P r(Z ≥ ) = .38
P-value = 2P r(K ≥ 28|H0 ) = 2P r Z ≥
4
8
d. He does not have ESP (because the P-value exceeds the significance level)

1

3.0321.065 or − 6. 6.2275.22 H0 : µ = 1.2 millimeters.5. P-value=P r(|T11 )| > ts ) = 2P r(t > 2.5.95 =4. 2 .32 + = 100 25 45 b.4 The differences Before-After are: d: 6 1 9 12 -1 9 ¯ and the standard deviation s=5. 2. 14. The sampled differences have mean d=6 The 99% confidence interval for the average decrease in the puzzle completion time is 99%C.6 a. Using table 4 we get .04.06 with t5.5.0639 and the 95% CI is of the form: (¯ x − y¯) ± t24.4653 5.04.02 < P − value < .05 x ¯s = 1.8475. 2 s2 We have: SE(¯x−¯y) = nsxx + nyy = . Open book tests make a difference. ts = (1.975 = 2. There is sufficient evidence .0314 < αso reject H0 . 5 14. There is a difference in the mean number of problems. and null hypothesis: H0 : x − y = 0 and alternative q hypothesis: Ha : x − y 6= 0. 99999999. sy = 3. to conclude that population average thickness is not 1.995 √ 6 So 6 D = 6 ± 8. 4.1 < x − y < −1.2275−1.87797 and test statistic: ts = SEx¯−¯y = 2. x ¯ = 15.975 = 2.Here df=25-1=24.4653) = 2 ∗ tcdf (2.10 Here nx = 24.72 2.0275) . sx = 2. t7. ny = 24. y¯ = 12.: s ¯ D = d = ±t5. 11) = .03864 √ 12 = 2. 12.2 α = .9 7 14. (¯ x−¯ y) Reject H0 if |ts | > t23.2 Ha : µ 6= 1.4 1.4653.02 < P − value < .975 SE(¯x−¯y) or 95%CI : −4 ± 2.3 Or: -2 seconds ≤D ≤ 14 seconds.0687 Conclusion: Reject H0 . s SE(¯x−¯y) = s2y s2x + = nx ny r 4.I.2) √s n = (.

717 ≤ x − y ≤ 18. 10 14.283 or 11. √ 95%CI : 15 ± 2. X2 ) = E(X1 X2 )E(X1 )E(X − 2) = π12 π1 π2 If we put it all together: V ar(Di ) = π1 (1 − π1 ) + π2 (1 − π2 ) − 2(π12 − π1 π2 ) K1 − K2 is the sum of n = 100 independent Di so according to the central limit theorem.752 7 + 11 b.8 14.75 = 15 ± 3.14 a. Here s2x nx = 2 and s2y ny 2 = 0. Finally ZD is just the bootstrapped approximation of standardized K1 − K2 so it is 3 .975 sp q 1 nx + 1 ny = (35 − 20) ± 2.05.975 SE(¯x−¯y) .3) + P r(K1 .20 a.1009(3.2|π1 = 0.283.3) = P r(Z > 59. Let Di = X1 − X2 . the standardized form of K1 . b. 40.987.4564) = 15 ± 3.4 ≤ x − y ≤ 18. Z1 .4238 (nx + ny − 2) 8 + 12 − 2 18 b. K1 − K2 is approximately normally distributed with mean and variance 100 times those of Di . c. The 95% confidence interval for x-y is of the form (¯ x − y¯) ± tdf. The 95% confidence interval x − y is (¯ x − y¯) ± t18. The pooled standard deviation is: s r r (nx − 1)Sx2 + (ny − 1)Sy2 7 ∗ 42 + 11 ∗ 32 221 sp = = = = 3.1788 1 + 0.4238)(0.41 p a. Let X1 = 1 if a randomly selected adult answers yes to the first question and 0 otherwise.8|π1 = 0. Let X2 = 1 if a randomly selected adult answers yes to the second question and 0 otherwise. 9 14. 21 d. 1 − β = P r(C|HA ) = P r(K1 > 59. α = P r(C|H0 ) = 0. will be approximately standard normal. because n = 100 is moderately large as long as 100π1 and 100(1 − π1 ) both are greater than 5.75 Welch’s formula: df= (2+0.75) = 12 22 0.8−30 √ ) = 0.613 or 11. K1 has a binomial distribution with mean µ1 = 100π1 and standard deviation 100π1 (1 − π1 ).6. Then: E(Di ) = π1 − π2 and V ar(Di ) = V ar(X1 + X2 ) = V ar(X1 ) + V ar(X2 ) − 2Cov(X1 X2 ) And Cov(X1 .

364 ≤ π1 − π2 ≤ −0. ZD = −4. e. Based on this random sample of 100 adults. An approximate 95% confidence interval π1 − π2 is: p (p1 (1 − p1 ) + p2 (1 − p2 ) − 2(p12 − p1 p2 )) π1 − π2 ± 1. An approximate (1 − α)100% confidence interval for π1 − π2 is: p CI : p1 − p2 ± z¯2 = 100(p1 (1 − p1 ) + p2 (1 − p2 ) − 2(p12 − p1 p2 )) f.approximately standard normal.96) g. there is sufficient evidence to support the hypothesis that there is a difference in the adult’s population’s response rate to these two questions at the 5% significance level.236 h. ZD defined in part d can be used to test this hypothesis.96 100 −0.7 is in the critical region. Yes. 4 . The critical region or rejection region is: C = (ZD : |ZD | > 1.