Introductory Econometrics
ECON3121
What is Econometrics?
Econometrics is about measuring economic relations.
What is the quantitative effect of reducing class size on
student achievement?
How does ano
Multiple Regression Analysis:
OLS Asymptotics
Small Sample/ Large Sample
Issues:
(1) Consistency
(2) Asymptotic Normality
(3) Asymptotic Efficiency
Consistency
Under the Gauss-Markov assumptions OL
Multiple Regression Analysis
y = 0 + 1x1 + 2x2 + . . . kxk + u
-allows us to explicitly control for many other
factors, and so allows ceteris paribus analysis
- allows to generalize functional relatio
Simple Linear Regression Model
y = 0 + 1x + u
Issues: Causality and Ceteris Paribus?
y = 1 x
if u=0
E(u) = 0
- Meaning:
In the population,
the average value of u, the error term, is 0.
- Assumption? N
CHAPTER 19
TEACHING NOTES
Students should read this chapter if you have assigned them a term paper. I used to allow
students to choose their own topics, but this is difficult in a first-semester cours
CHAPTER 18
SOLUTIONS TO PROBLEMS
18.1 With zt1 and zt2 now in the model, we should use one lag each as instrumental variables, zt- 1,1
and zt- 1,2. This gives one overidentifying restriction that can
CHAPTER 18
TEACHING NOTES
Several of the topics in this chapter, including testing for unit roots and cointegration, are now
staples of applied time series analysis. Instructors who like their course
CHAPTER 17
SOLUTIONS TO PROBLEMS
17.1 (i) Let m0 denote the number (not the percent) correctly predicted when yi = 0 (so the
prediction is also zero) and let m1 be the number correctly predicted when
CHAPTER 17
TEACHING NOTES
I emphasize to the students that, first and foremost, the reason we use the probit and logit models
is to obtain more reasonable functional forms for the response probability
CHAPTER 16
SOLUTIONS TO PROBLEMS
16.1 (i) If 1 = 0 then y1 = 1z1 + u1, and so the right-hand-side depends only on the exogenous
variable z1 and the error term u1. This then is the reduced form for y1.
CHAPTER 16
TEACHING NOTES
I spend some time in Section 16.1 trying to distinguish between good and inappropriate uses of
SEMs. Naturally, this is partly determined by my taste, and many applications f
CHAPTER 15
SOLUTIONS TO PROBLEMS
15.1 (i) It has been fairly well established that socioeconomic status affects student performance.
The error term u contains, among other things, family income, which
Multiple Regression Analysis: Further Issues
1. Data Scaling
2. Functional Form - Logarithmic/Quadratic/Interaction
terms
(Nonlinear functions of x and y,
but still linear in the parameters)
3. Goodne
Binary (Dummy) Variable
A Single Dummy Independent Variable
y = 0 + 0D + 1x + u
e.g. wage= 0 + 0female + 1educ + u
Comparison-of-means
Simple regression on a constant and a dummy variable is a
straigh
Introductory Econometrics
Lecture 1 - Introduction & Review of Statistics
Jin Yan
The Chinese University of Hong Kong
September 11, 2017
Lecture 1
Introductory Econometrics
September 11, 2017
1 / 60
W
Introductory Econometrics
Lecture 4 - Multiple Regression (I)
Jin Yan
The Chinese University of Hong Kong
September 28, 2017
Lecture 4
Multiple Regression
September 28, 2017
1 / 67
Announcements
Next
Introductory Econometrics
Lecture 3 - Simple Regression (II)
Jin Yan
The Chinese University of Hong Kong
September 25, 2017
Lecture 3
Simple Regression
September 25, 2017
1 / 43
Notice
Problem Set 1:
Introductory Econometrics
Lecture 2 - Simple Regression (I)
Jin Yan
The Chinese University of Hong Kong
September 18 , 2017
Lecture 2
Introductory Econometrics
September 18 , 2017
1 / 44
Notice
Proble
Problem Set 1
Introdutory Eonometris
Eon 3121B, Fall 2017
Due: In lass on Monday, September 25, 2017
(no late problem set aepted)
*Please answer as suintly as possible while still fully answering the
Introduction to Stata
ECON 3121B Tutorial
DOWNLOAD
data and code file from:
Blackboard-course content-tutorials
1
The Four Windows of Stata
Review
Variable Window
Results Window
Command Window
2
The S
ECON3121 Miscellaneous Material (Not for the Examination)
A. Sampling Variances of the OLS estimator,
Interceptin SLR
n
P
21
V ar( ^ 0 ) =
n i=1
n
P
(xi
x2
i
(1)
x)2
i=1
(Proof) Conditioning on x,
V
ECON3121, Understanding STATA
The aim of this lecture is to introduce very basic STATA commands and to give opportunities
for you to practice econometrics with actual data. You are supposed to learn t
Real basic mathematical/statistical knowledge:
This material is based on Professor Chong previous lecture note for this class. You may
s
also want to study the appendix of the Wooldridge book. Wheneve
Fixed Effects Estimation
FE or FD?
Assumptions for Pooled OLS using FE
Dummy Variable Regression
Unobserved Heterogeneity
Pooled OLS with
Quasi-demeaned Data
Even in the case that unobserved heteroge
Panel Data Analysis
Types of data
Cross-sectional data:
obtained by random sampling at a given point in time
Time series data:
have a separate observation for each time period
Cross-sectional data +
Heteroskedasticity
Homoskedasticity - Var(u|x) = 2
Heteroskedastic Case
Heteroskedasticity with a single regressor
= + ( xi x ) ui
1
1
(xi x )2
(xi x )2 i2
Var (1 ) =
2
( ( xi x ) ) 2
Heteroskedasti
CHAPTER 15
TEACHING NOTES
When I wrote the first edition, I took the novel approach of introducing instrumental variables as
a way of solving the omitted variable (or unobserved heterogeneity) problem
CHAPTER 14
SOLUTIONS TO PROBLEMS
2
14.1 First, for each t > 1, Var( uit) = Var(uit ui,t- 1) = Var(uit) + Var(ui,t- 1) = 2 u , where we use
the assumptions of no serial correlation in cfw_ut and consta
CHAPTER 7
TEACHING NOTES
This is a fairly standard chapter on using qualitative information in regression analysis, although
I try to emphasize examples with policy relevance (and only cross-sectional