Scribd
Upload
88 views10 pages

Z Test (Standard Normal Distribution) (N 30) : 1) Confidence Interval For Mean

The document summarizes common statistical tests used for hypothesis testing, including the z-test, t-test, F-test, and chi-square test. It provides formulas and explanations for confidence intervals and hypothesis tests related to means, proportions, standard deviations, and the independence of attributes using these distributions. Degrees of freedom and test statistics are defined for each test.

Uploaded by

Aniruddha Phalak
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
88 views10 pages

Z Test (Standard Normal Distribution) (N 30) : 1) Confidence Interval For Mean

The document summarizes common statistical tests used for hypothesis testing, including the z-test, t-test, F-test, and chi-square test. It provides formulas and explanations for confidence intervals and hypothesis tests related to means, proportions, standard deviations, and the independence of attributes using these distributions. Degrees of freedom and test statistics are defined for each test.

Uploaded by

Aniruddha Phalak
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Z test (standard normal distribution) (n > 30)

1) Confidence interval for mean


𝜎 𝜎 𝑠 𝑠
(𝑥̅ − 𝑧𝛼 , 𝑥̅ + 𝑧𝛼 ) 𝑜𝑟 (𝑥̅ − 𝑧𝛼 , 𝑥̅ + 𝑧𝛼 )
√𝑛 √𝑛 √𝑛 √𝑛

2) One sample test for mean 𝐻0 : 𝑥̅ = 𝜇


𝑥̅ − 𝜇 𝑥̅ − 𝜇
𝑍= 𝑜𝑟 𝑍 =
𝜎⁄√𝑛 𝑠⁄√𝑛

3) Two sample test for mean 𝐻0 : 𝑥̅1 = 𝑥̅2


i) population SD known
𝑥̅1 − 𝑥̅2
𝑍=
𝜎2 𝜎2
√ 1 + 2
𝑛1 𝑛2

ii) population SD unequal and unknown


𝑥̅1 − 𝑥̅2
𝑍=
𝑠2 𝑠2
√ 1+ 2
𝑛1 𝑛2

iii) population SD equal and unknown


𝑥̅1 − 𝑥̅2
𝑍=
𝑠12 𝑠22
√ +
𝑛2 𝑛1

4) Confidence interval for proportion


𝑝̂ 𝑞̂ 𝑝̂ 𝑞̂
( 𝑝̂ − 𝑧𝛼 √ , 𝑝̂ + 𝑧𝛼 √ )
𝑛 𝑛

Page 1 of 2
5) One sample test for proportion 𝐻0 : 𝑝̂ = 𝑃
𝑝̂ − 𝑃
𝑍=
√𝑃𝑄
𝑛

6) Two sample test for proportion 𝐻0 : 𝑝1 = 𝑝2


𝑝1 − 𝑝2
𝑍=
1 1
√𝑃𝑄 ( + )
𝑛1 𝑛2
𝑛1 𝑝1 +𝑛2 𝑝2
If P is unknown then use 𝑃 =
𝑛1 +𝑛2

7) One sample test for SD 𝐻0 : 𝑠 = 𝜎


𝑠−𝜎
𝑍=
𝜎⁄√2𝑛

8) Two sample test for SD 𝐻0 : 𝑠1 = 𝑠2


i) population SD known
𝑠1 − 𝑠2
𝑍=
1 1
𝜎√ +
2𝑛1 2𝑛2
ii) population SD unknown
𝑠1 − 𝑠2
𝑍=
𝑠2 𝑠2
√ 1 + 2
2𝑛2 2𝑛1

Page 2 of 2
t test (Student’s t-distribution) (n < 30)

1) Confidence interval for mean


𝑠 𝑠
(𝑥̅ − 𝑡𝛼 , 𝑥̅ + 𝑡𝛼 )
√𝑛 − 1 √𝑛 − 1

2) One sample test for mean 𝐻0 : 𝑥̅ = 𝜇


𝑥̅ − 𝜇
𝑡=
𝑠⁄√𝑛 − 1
𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = 𝑛 − 1

3) Two sample test for mean


i) samples are independent and 𝑛1 ≠ 𝑛2
𝐻0 : 𝑥̅1 = 𝑥̅2
𝑥̅1 − 𝑥̅2
𝑡=
𝑛 𝑠 2 + 𝑛2 𝑠22 1 1
√( 1 1 ) ( + )
𝑛1 + 𝑛2 − 2 𝑛1 𝑛2
𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = 𝑛1 + 𝑛2 − 2

i) samples are independent and 𝑛1 = 𝑛2 = 𝑛

𝐻0 : 𝑥̅1 = 𝑥̅2
𝑥̅1 − 𝑥̅2
𝑡=
𝑠 2 + 𝑠22
√( 1
𝑛−1 )
𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = 2𝑛 − 2

ii) samples are dependent 𝐻0 : 𝑑̅ = 0


𝑑̅
𝑡=
𝑠⁄√𝑛 − 1
𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = 𝑛 − 1
1 2
Here 𝑑̅ = 𝑥̅ − 𝑦̅ , 𝑑𝑖 = 𝑥𝑖 − 𝑦𝑖 , 𝑎𝑛𝑑 𝑠 2 = ∑𝑛𝑖=1(𝑑𝑖 − 𝑑̅ )
𝑛
Page 1 of 3
F test (Snedecor’s F-distribution) (n < 30)

1) Two sample test for variance 𝐻0 : 𝜎12 = 𝜎22


𝜎12
𝐹= 2 ; (𝜎12 ≥ 𝜎22 )
𝜎2
𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 𝑜𝑓 𝑛𝑢𝑚𝑒𝑟𝑎𝑡𝑜𝑟 = 𝑛1 − 1
𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 𝑜𝑓 𝑑𝑒𝑛𝑜𝑚𝑖𝑛𝑎𝑡𝑜𝑟 = 𝑛2 − 1
𝑛1 𝑠12 𝑛2 𝑠22
Here 𝜎12 = 𝑎𝑛𝑑 𝜎22 =
𝑛1 −1 𝑛2 −1

𝝌𝟐 test (Chi-square distribution)

1) Goodness of Fit
𝐻0 : 𝑂𝑏𝑠𝑒𝑟𝑣𝑒𝑑 𝑣𝑎𝑙𝑢𝑒𝑠 𝑓𝑜𝑙𝑙𝑜𝑤 𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
𝑛
2
(𝑂𝑖 − 𝐸𝑖 )2
𝜒 =∑
𝐸𝑖
𝑖=1

𝐺𝑒𝑛𝑒𝑟𝑎𝑙𝑙𝑦, 𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = 𝑛 − 1


Degree of freedom = Number of observations – Number of independent
constraints
2) Independence of Attributes
𝐻0 : 𝐴𝑡𝑡𝑟𝑖𝑏𝑢𝑡𝑒𝑠 𝑎𝑟𝑒 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡
𝑚 𝑛 2
(𝑂𝑖𝑗 − 𝐸𝑖𝑗 )
𝜒2 = ∑ ∑
𝐸𝑖𝑗
𝑖=1 𝑗=1

𝐺𝑒𝑛𝑒𝑟𝑎𝑙𝑙𝑦, 𝑑𝑒𝑔𝑟𝑒𝑒 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = (𝑚 − 1)(𝑛 − 1)


𝑅𝑖 𝐶𝑗
Here 𝐸𝑖𝑗 = 𝑎𝑛𝑑 𝑁 = ∑𝑚 𝑛
𝑖=1 𝑅𝑖 = ∑𝑗=1 𝐶𝑗
𝑁
Page 2 of 3
Yate’s Correction:
For 2x2 contingency table,
2 2 2
(|𝑂𝑖𝑗 − 𝐸𝑖𝑗 | − 0.5)
𝜒2 = ∑ ∑
𝐸𝑖𝑗
𝑖=1 𝑗=1

Page 3 of 3

You might also like