Title: Regression with a Single Regressor: Hypothesis Tests and Confidence Intervals
1Chapter 5
- Regression with a Single Regressor Hypothesis
Tests and Confidence Intervals
2Regression with a Single Regressor Hypothesis
Tests and Confidence Intervals(SW Chapter 5)
3But first a big picture view (and review)
4Object of interest ?1 in,
5Hypothesis Testing and the Standard Error of
(Section 5.1)
6(No Transcript)
7Formula for SE( )
8(No Transcript)
9Summary To test H0 ?1 ?1,0 v. H1 ?1 ?
?1,0,
10Example Test Scores and STR, California data
11(No Transcript)
12Confidence Intervals for ?1(Section 5.2)
13(No Transcript)
14A concise (and conventional) way to report
regressions
15OLS regression reading STATA output
16Summary of Statistical Inference about ?0 and ?1
17Regression when X is Binary(Section 5.3)
18Interpreting regressions with a binary regressor
19(No Transcript)
20Summary regression when Xi is binary (0/1)
21Heteroskedasticity and Homoskedasticity, and
Homoskedasticity-Only Standard Errors (Section
5.4)
22(No Transcript)
23Homoskedasticity in a picture
24Heteroskedasticity in a picture
25A real-data example from labor economics
average hourly earnings vs. years of education
(data source Current Population Survey)
26The class size data
27So far we have (without saying so) assumed that u
might be heteroskedastic.
28What if the errors are in fact homoskedastic?
29(No Transcript)
30We now have two formulas for standard errors for
31Practical implications
32Heteroskedasticity-robust standard errors in
STATA
33The bottom line
34Some Additional Theoretical Foundations of OLS
(Section 5.5)
35Further Questions
36The Extended Least Squares Assumptions
37Efficiency of OLS, part I The Gauss-Markov
Theorem
38The Gauss-Markov Theorem, ctd.
39Efficiency of OLS, part II
40Some not-so-good thing about OLS
41Limitations of OLS, ctd.
42Inference if u is Homoskedastic and Normal the
Student t Distribution (Section 5.6)
43(No Transcript)
44(No Transcript)
45Practical implication
46Summary and Assessment (Section 5.7)