Scribd
Upload
12 views6 pages

Lecture 2 - Regression

This document provides an introduction to simple and multiple linear regression analysis. It defines key terms like the response variable, predictor variables, and regression slope. It explains how to interpret the regression coefficients (b0 and b1) and how b1 indicates the strength of the relationship between the independent and dependent variables. It discusses performing hypothesis tests and constructing confidence intervals for the regression slope b1. Finally, it reviews concepts like residuals, the F-distribution, R-squared, and analysis of variance (ANOVA) testing for assessing the fit and significance of the regression model.

Uploaded by

d5zb58j5dm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
12 views6 pages

Lecture 2 - Regression

This document provides an introduction to simple and multiple linear regression analysis. It defines key terms like the response variable, predictor variables, and regression slope. It explains how to interpret the regression coefficients (b0 and b1) and how b1 indicates the strength of the relationship between the independent and dependent variables. It discusses performing hypothesis tests and constructing confidence intervals for the regression slope b1. Finally, it reviews concepts like residuals, the F-distribution, R-squared, and analysis of variance (ANOVA) testing for assessing the fit and significance of the regression model.

Uploaded by

d5zb58j5dm
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Quants – Lecture 2 – Regression

Introduction
 Response variable or dependent: variable being predicted.
 Predictor variables or independent variables: variables being used to predict the value of
the dependent variable.
 Simple (single) linear regression: a regression analysis for which any one unit change in
the independent variable, x, is assumed to result in a change in the dependent variable,
y.
 Multiple linear regression: a regression analysis involving two or more independent
variables.

What do b0, b1 mean?


 Interpretation of b1:
o If the length of a driving assignment were 1 unit (mile) longer, the mean
travel time for that driving assignment would be 0.0678 units (0.0678 hours,
or approximately 4 minutes) longer.
 Interpretation of b0:
o If the driving distance for a driving assignment was 0 units (0 miles), the
mean travel time would be 1.2739 units (1.2739 hours, or approximately 76
minutes) – not a useful meaning.

Inference for Regression


 Our regression model is just based on one sample. It might change from sample to
sample.
 What can we say about the actual relationship between variables in the population?
 We need the sampling distribution of the slope (B1) to make inferences/
 (Why do we focus on b1, i.e., slope, and not b0?
o Because b1 shows the existence of a relationship between variables, not b0.

Confidence Interval for the Regression Slope


 When the assumptions are met, we can find a confidence interval for β 1 from

Where the critical value t ¿ depends on the confidence level and has n-2 degrees of freedom.

Use technology (Excel) to calculate SE (b1) and confidence interval for you.
Assumptions and Conditions
1. Make sure you have quantitative variables x, y.
2. Draw a scatterplot and check if there is a linear relationship between variables.
3. Fit a regression line and calculate residuals (e). Make a scatterplot for residuals against
variable x or the predicted values. This plot should have no patterns.
Hypothesis Test for the Regression Slope: t-test for the Regression Slope
 When the assumptions and conditions are met, we can test the hypothesis H0: b1 = 0 vs.
HA: B1 does not = 0 (or a one-sided alternative hypothesis) using the standardized
estimated regression slope
b 1−beta 1
 t=
SE(b1)
 Which follows a student’s t-model with n-2 degrees of freedon, We can use the t-model
to find the P-value of the test.
ANOVA and the F-Statistic
 Analysis of variance (ANOVA) for regression is another test for beta1 (b1).
 But instead of stating the test in terms of the parameter, we restate it in terms of the
model.
 Is the regression model worthwhile?
o (i.e., does the predictor variable contain useful information about the response
variable?)

F-Distribution
(review)
1. The curve is
not
symmetrical
but skewed to
the right.
2. There is a
different curve
for each set of
dfs.
3. As the
degrees of freedom for the
numerator and for the denominator
get larger, the curve approximates the
normal.
Variation in the Model and R-Squared
 All regression models fall somewhere between the two extremes of zero correlation or
perfect correlation plus or minus 1.
 We consider the square of the correlation co-efficient r to get r-squared which is a value
between 0 and 1.
 R-squared: the fraction of the data’s variation accounted for by the model.
 1-r-squared: the fraction of the original variation left in the residuals.
 Two ways to calculate r-squared
o Square the correlation of r (r^2)

o
o

You might also like