Business Statistics and Analysis3
课件来源于Rice University的Business Statistics and Analysis
四，Linear Regression for Business Statistics
Week 1 – Regression Analysis
1. Introducing Linear Regression Building the Model
What is Linear Regression?
Linear regression attempts to fit a linear relation between a variable of interest and a set of variables that may be related to the variable of interest.
1 2 3 4 5 6 7 8 9 10 11 12 

Overview of Regression:
Simple Regression:
A regression with only one explanatory or X variable
Multiple Regression:
A regression with two or more explanatory or X variables
 Modeling Developing a regression model
 Estimation Using software to estimate the model
 Inference Interpreting the estimated regression model
 Prediction Making predictions about the variable of interest
2. Errors, Residuals and Rsquare
Regression is a process that has errors
 Residuals and Errors
 Rsquare: A “goodness of fit” measure.
Why do we have errors in the regression model ?
 Omitted variables.
 Functional relationship between the Y and X variables.
 The theory of regression analysis is based on certain assumptions about these errors.
 We have differentiated between the β’s and the b’s.
 In common usage people tend to mix these notations.
 However, as long as you understand the conceptual difference, you should be ok.
Week 2 – Hypothesis Testing and Goodness of Fit
1. Hypothesis Testing in a Linear Regression
2. Lesson Hypothesis Testing in a Linear Regression ‘pvalues’
 The tcutoff approach
 The pvalue approach
 The confidence interval approach
3. ‘Goodness of Fit’ measures Rsquare and Adjusted Rsquare
4. Categorical Variables in a Regression Dummy Variables
1 2 3 4 5 6 

Week 3 – Dummy Variables, Multicollinearity
1. Dummy Variable Regression – Extension to Multiple Categories
We need one dummy variable less than the number of categories. Incorrect to have three dummies REGA, REGB and REGC.