Chapter 6

Introduction to
Multiple Linear
Regression

Outline
Omitted variable bias
Causality and regression analysis, affected by multicoliniarity
Multiple regression and OLS
Measures of fit
Sampling distribution of the OLS estimator
Hypothesis tests and confidence intervals for a single
coefficient
Joint hypothesis tests on multiple coefficients
How to decide what variables to include in a regression model?
2

Omitted Variable Bias

(SW Section 6.1)

Omitted variable bias, ctd.

Omitted variable bias, ctd.

The omitted variable bias formula:

Digression on causality and

regression analysis

Ideal Randomized Controlled

Experiment

Ideal: subjects all follow the treatment protocol perfect

compliance, no errors in reporting, etc.!
Randomized: subjects from the population of interest are
randomly assigned to a treatment or control group (so
there are no confounding factors)
Controlled: having a control group permits measuring the
differential effect of the treatment
Experiment: the treatment is assigned as part of the
experiment: the subjects have no choice, so there is no
reverse causality in which subjects choose the treatment
they think will work best.
8

Back to class size:

10

Return to omitted variable bias

11

The Population Multiple Regression

Model (SW Section 6.2)

12

Interpretation of coefficients in
multiple regression

13

14

The OLS Estimator in Multiple

Regression (SW Section 6.3)

15

Example: the California test score

data

16

Multiple regression in STATA

17

Measures of Fit for Multiple

Regression (SW Section 6.4)

18

SER and RMSE

19

R2 and R

20

R2 and R , ctd.

21

Measures of fit, ctd.

22

The Least Squares Assumptions for

Multiple Regression (SW Section 6.5)

23

24

The Sampling Distribution of the

OLS Estimator (SW Section 6.6)

25

Imperfect multicollinearity

26

Hypothesis Tests and Confidence Intervals

for a Single Coefficient in Multiple
Regression (SW Section 7.1)

27

Example: The California class size data

28

Standard errors in multiple

regression in STATA

29

IN DEPHT TESTING: Tests of Joint Hypotheses

(SW Section 7.2)

30

Tests of joint hypotheses, ctd.

31

Test the coefficients one at a time?

32

Suppose t1 and t2 are independent (for

this calculation).

33

34

The F-statistic

35

36

Computing the p-value using the

F-statistic:

37

F-test example, California class size

data:

38

Summary: testing joint hypotheses

39

Testing Single Restrictions on

Multiple Coefficients (SW Section 7.3)

40

Testing single restrictions on

multiple coefficients, ctd.

41

Method 2: Perform the test directly

42

Confidence Sets for Multiple

Coefficients (SW Section 7.4)

43

Joint confidence sets ctd.

44

45

Confidence set based on inverting

the F-statistic

46

Digression about measures of fit

47

Digression on presentation of
regression results

48

Summary: Multiple Regression

49