COMM 215 BUSINESS STATISTICS (Bowerman 8th Edition)

Chapter 2 Descriptive Statistics: Tabular and Binomial Probability Function: 𝑃(𝑥) =

Graphical Presentations. 𝑛!
𝑝 𝑥 𝑞𝑛−𝑥
𝑥!(𝑛−𝑥)!

𝑎𝑝𝑝𝑟𝑜𝑥𝑖𝑚𝑎𝑡𝑒 𝑐𝑙𝑎𝑠𝑠 𝑙𝑒𝑛𝑔𝑡ℎ

Expected Value for the Binomial Distribution: 𝜇𝑥 =
𝑙𝑎𝑟𝑔𝑒𝑠𝑡 𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑚𝑒𝑛𝑡 − 𝑠𝑚𝑎𝑙𝑙𝑒𝑠𝑡 𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑚𝑒𝑛𝑡
= 𝑛𝑝
𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑙𝑎𝑠𝑠𝑒𝑠
Variance for the Binomial Distribution: 𝜎𝑥2 = 𝑛𝑝𝑞
Chapter 3 Descriptive Statistics: Quantitative
Chapter 6 Continuous Random Variables
Interquartile Range: 𝐼𝑄𝑅 = 𝑄3 − 𝑄1
The Standard Normal Distribution:
Sample Variance:
𝑥−𝜇
∑𝑛𝑖=1(𝑥𝑖 − 𝑥̅ )2 𝑧=
𝑠2 = 𝜎
𝑛−1
Chapter 7 Sampling Distribution
𝑛
1 (∑𝑛𝑖=1 𝑥𝑖 )2
𝑠2 = [∑ 𝑥𝑖2 – ] The Sampling Distribution of 𝑥̅
𝑛−1 𝑛
𝑖=1 𝜇𝑥̅ = 𝜇
𝜎
𝜎𝑥̅ =
Chapter 4 Probability √𝑛
Sample error 𝑝̂ of :
𝑁 𝑁!
𝜇𝑝̂ = 𝑝
Counting Rule for Combinations: ( ) =
𝑛 𝑛!(𝑁−𝑛)!
𝑝(1 − 𝑝)
Addition Rule: 𝑃(𝐴 ∪ 𝐵) = 𝑃(𝐴) + 𝑃(𝐵) − 𝜎𝑝̂ = √
𝑛
𝑃(𝐴 ∩ 𝐵)
Chapter 8 Confidence Intervals
Conditional Probability: 𝑃(𝐴|𝐵) = z-Based confidence intervals for Population Mean:
𝜎 Known
𝑃(𝐴 ∩ 𝐵)⁄𝑃(𝐵) 𝜎
𝑥̅ ± 𝑧𝛼/2
The Multiplication Rule:𝑃 (𝐴 ∩ 𝐵) = 𝑃(𝐵)𝑃(𝐴|𝐵) √𝑛
t-Based Confidence Intervals for a Population
Chapter 5 Discrete Random Variables Mean: 𝜎 Unknown
𝑠
𝑥̅ ± 𝑡𝛼/2
The Expected Value of a Discrete Random √𝑛
Sample Size for a Confidence Interval for
Variable:
𝜇: 𝜎 Known
𝜇𝑥 = ∑ 𝑥𝑝(𝑥)
𝐴𝑙𝑙 𝑥 𝑧𝛼/2 𝜎 2
𝑛= ( )
𝐸
Variance of a Discrete Random Variable:
Confidence Interval for the Proportion:
𝜎𝑥2 = ∑ (𝑥 − 𝜇𝑥 )2 𝑝(𝑥)
𝐴𝑙𝑙 𝑥 𝑝̂ (1 − 𝑝̂ )
𝑝̂ ± 𝑧𝛼/2 √
𝑛
Number of ways to arrange x successes among n
trials:

𝑁 𝑛!
( )=
𝑛 (𝑛
𝑥! − 𝑥)!

1
Chapter 9 Hypothesis Testing 𝑆𝑆𝐸
Standard error of the estimate 𝑠 = √𝑛−𝑘−1
𝑥̅ −𝜇0
z-test for the mean 𝑧 = 𝜎
⁄ 𝑛 𝑆𝑆𝑅
√ Coefficient of Determination: 𝑅 2 = 𝑟 2 = 𝑆𝑆𝑇
𝑥̅ −𝜇0
t-test for the mean 𝑡 = 𝑠 F-test for the simple linear regression model
⁄ 𝑛
√
𝑆𝑆𝑅⁄
𝐹= 𝑘
𝑝̂−𝑝0 𝑆𝑆𝐸⁄
z-test for proportion 𝑧 = 𝑛−𝑘−1
𝑝 (1−𝑝0 )
√ 0
𝑛

Simple regression estimator for the standard error

Chapter 12 Goodness-of-Fit Tests of the slope:
Chi-Square goodness-of-fit test statistic 𝑠
𝑠𝑏1 =
√𝑆𝑆𝑥𝑥
𝑘
2
(𝑓𝑖 − 𝐸𝑖 )2
𝜒 =∑ 𝑏1 − 𝛽1
𝐸𝑖 Test of hypothesis for the slope: 𝑡 =
𝑖=1 𝑠𝑏1
Chi-square contingency test statistic df= n-k-1
𝑟 𝑐 2
2
(𝑓𝑖𝑗 − 𝐸̂𝑖𝑗 ) Confidence Interval for the mean value of y
𝜒 = ∑∑
𝐸̂𝑖𝑗
𝑖=1 𝑗=1
with 𝑑𝑓 = (𝑟 − 1)(𝑐 − 1) 1 (𝑥0 − 𝑥̅ )2
𝑦̂ ± 𝑡𝛼/2 𝑠 √ +
𝑛 𝑆𝑆𝑥𝑥
Chapter 13 Simple Linear Regression Analysis
Prediction interval for an individual value of y
Sample correlation coefficient
Simple linear regression model:
𝑦 = 𝛽0 + 𝛽1 𝑥 + ℇ 1 (𝑥0 − 𝑥̅ )2
𝑦̂ ± 𝑡𝛼/2 𝑠 √1 + +
𝑛 𝑆𝑆𝑥𝑥
Least Squares point estimate of the slope 𝛽1
𝑆𝑆𝑥𝑦 Chapter 14 Multiple Regression
𝑏1 = 𝑤ℎ𝑒𝑟𝑒
𝑆𝑆𝑥𝑥
The multiple regression model:
𝑆𝑆𝑥𝑦 = ∑(𝑥𝑖 − 𝑥̅ ) (𝑦𝑖 − 𝑦̅) 𝑦 = 𝛽0 + 𝛽1 𝑥1 + 𝛽2 𝑥2 + ⋯ + + 𝛽𝑘 𝑥𝑘 + ℇ
(∑ 𝑥𝑖 )(∑ 𝑦𝑖 )
= ∑ 𝑥𝑖 𝑦𝑖 −
𝑛
Standard error of the estimate:
(∑ 𝑥𝑖 )2 𝑆𝑆𝐸
𝑆𝑆𝑥𝑥 = ∑(𝑥𝑖 − 𝑥̅ )2 = ∑ 𝑥𝑖2 − 𝑠=√ = √𝑀𝑆𝐸
𝑛 𝑛−𝑘−1
Least squares point estimate of the y-intercept 𝛽0
Multiple coefficient of determination:
𝑆𝑆𝑅
𝑏0 = 𝑦̅ − 𝑏1 𝑥̅ 𝑅 2 = 𝑟 2 = 𝑆𝑆𝑇

Sum of squares residuals (Sum of squares error) An F-test for the linear regression model:
Total variation SST = ∑(𝑦𝑖 − 𝑦̅)2 𝑆𝑆𝑅⁄
𝐹= 𝑘
Explained variation SSR = ∑(𝑦̂𝑖 − 𝑦̅)2 𝑆𝑆𝐸⁄
Unexplained variation SSE = ∑(𝑦𝑖 − 𝑦̂𝑖 )2 𝑛−𝑘−1
SSE = ∑ 𝑦𝑖2 − 𝑏0 ∑ 𝑦𝑖 − 𝑏1 ∑ 𝑥𝑖 𝑦𝑖