Professional Documents
Culture Documents
Chapter 10
Hypothesis
Testing: Additional
Topics
di = xi − yi
• Assumptions:
– Both Populations Are Normally Distributed
Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 4
Test Statistic: Dependent Samples
Population Means, Dependent Samples
𝑑
Where 𝑡 = 𝑠 has 𝑛 − 1 d. f.
𝑑
𝑛
Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 6
Matched Pairs Example
• Assume you send your salespeople to a “customer
service” training workshop. Has the training made a
difference in the number of complaints? You collect the
following data: σ𝑑 𝑖
𝑑=
𝑛
= −4.2
σ(𝑑𝑖 − 𝑑)2
𝑠𝑑 =
𝑛−1
= 5.67
𝛼 = .05 𝑑 = −4.2
i. e. , i. e. , i. e. ,
𝐻0 : 𝜇𝑥 – 𝜇𝑦 ≥ 0 𝐻0 : 𝜇𝑥 – 𝜇𝑦 ≤ 0 𝐻0 : 𝜇𝑥 – 𝜇𝑦 = 0
𝐻1 : 𝜇𝑥 – 𝜇𝑦 < 0 𝐻1 : 𝜇𝑥 – 𝜇𝑦 > 0 𝐻1 : 𝜇𝑥 – 𝜇𝑦 ≠ 0
α = 0.05
df = 21 + 25 − 2 = 44
Critical Values: t = ±2.0154
𝑥lj − 𝑦lj
𝑡=
2
𝑠𝑥2 𝑠𝑦
+
𝑛𝑋 𝑛𝑌
2
𝑠𝑥2 𝑠𝑦2
+
𝑛𝑥 𝑛𝑦
𝑣= 2 2
Where t has degrees of freedom: 𝑠𝑥2 𝑠𝑦2
𝑛𝑥 𝑛𝑦
+
𝑛𝑥 − 1 𝑛𝑦 − 1
Assumptions:
Both sample sizes are large,
𝑛𝑃 1 − 𝑃 > 5
is a z value:
𝑝Ƹ𝑥 − 𝑝Ƹ𝑦
𝑧=
𝑝Ƹ0 1 − 𝑝Ƹ0 𝑝Ƹ 1 − 𝑝Ƹ0
+ 0
𝑛𝑥 𝑛𝑦
𝑛𝑥 𝑝Ƹ𝑥 + 𝑛𝑦 𝑝Ƹ𝑦
𝑝Ƹ0 =
Where 𝑛𝑥 + 𝑛𝑦
𝑝Ƹ𝑀 − 𝑝Ƹ𝑊
𝑧=
𝑝Ƹ0 1 − 𝑝Ƹ0 𝑝Ƹ 1 − 𝑝Ƹ0
+ 0
𝑛1 𝑛2
.50 − .62
=
.549 1 − .549 .549 1 − .549
+
72 50 Decision: Do not reject H0
Conclusion: There is not
= −1.31
significant evidence of a
difference between men
Critical Values = ±1.96
and women in proportions
For α = .05 who will vote yes.
Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 36
Section 10.4 Tests of Equality of Two
Variances
• Goal: Test hypotheses about two
population variances
𝐻0 : 𝜎𝑥2 ≥ 𝜎𝑦2
𝐻1 : 𝜎𝑥2 < 𝜎𝑦2 Lower-tail test
𝐻0 : 𝜎𝑥2 ≤ 𝜎𝑦2
𝐻1 : 𝜎𝑥2 > 𝜎𝑦2 Upper-tail test
𝐻0 : 𝜎𝑥2 = 𝜎𝑦2
𝐻1 : 𝜎𝑥2 ≠ 𝜎𝑦2 Two-tail test
𝑠𝑥2
𝐹= 2
𝑠𝑦
Degrees of Freedom:
• Numerator Fnx −1, ny −1,
(NYSE has the larger 2
standard deviation):
=𝐹 10 = 2.03
– 𝑛𝑥 − 1 = 21 − 1 = 20 d.f. 20,24,0.
2
• Denominator:
– 𝑛𝑦 − 1 = 25 − 1 = 24 d.f.
Copyright © 2020 Pearson Education Ltd. All Rights Reserved. Slide - 44
F Test: Example Solution (2 of 2)
• The test statistic is: 𝐻0 : 𝜎𝑥2 = 𝜎𝑦2
𝐻1 : 𝜎𝑥2 ≠ 𝜎𝑦2
For variances…
• F test for two variances:
– Data data analysis 𝐹 − test: two sample for variances