More on FGLS and How to Test for Heteroskedasticity
Generalized Least Squares and Feasible GLS
Heres a little more detail on how to perform GLS/FGLS:
Suppose you have a standard multivariate model such as
y i 0 1 x 1i 2 x 2i i
which satisfies all of the s
Takashi Yamano
Lecture Notes on Advanced Econometrics
Lecture 4: Multivariate Regression Model in Matrix Form
In this lecture, we rewrite the multiple regression model in the matrix form. A general
multiple-regression model can be written as
for i = 1, ,n
14.32 Recitation 2012-02-10
Miikka Rokkanen
February 11, 2012
1
Accessing Stata
1.1
Athena Clusters
Open the Athena shell and type add stata. Then type xstata to load the Statas GUI environment.
1.2
PC
You need to get access to Athena via SSH. For this pu
14.32 SPRING 2015
1. TU Feb. 3 Review pset out
2. TH Feb. 5 Pset 1 out
F Feb. 6 Review pset due
3. TU Feb. 10
4. TH Feb. 12
No class on TU Feb. 17
5. TH Feb. 19 Pset 1 due; Pset 2 out
6. TU Feb. 24
7. TH Feb. 26
8. TU March 3
9. TH March 5 Pset 2 due ; Ps
MIT 14.32
J. Angrist
Spring 2015
Course Outline and Study Guide
Part I
A. Statistical tools
1. Expectation and moments
B. Review of statistical inference
2. Sampling distributions and inference
3. Confidence intervals
C. Analysis and interpretation of ran
Econometrics
MIT (14.32)
Spring 2015
J. Angrist ([email protected])
Luu Nguyen ([email protected])
Stephanie Cheng ([email protected])
Our aim is to help you to understand modern applied econometric methods and to foster the skills needed
to plan and execute your o
Notes on the Bootstrap
Rachael Meager
November 3, 2015
The bootstrap, invented by the statistician Bradley Efron in 1979, is a resampling
technique for characterising sampling uncertainty. The core idea is that if we have iid
data, we can take repeated sa
Econometric Techniques for Binary or Limited Outcome Variables
14.32 Supplementary Lecture Note.
Rachael Meager
October 22, 2015
1
Binary Outcome Variables
Suppose we want to know how a set of covariates X influences a binary outcome y. We
will be primari
14.32 PSET 1
Rachael Meager
September 16, 2015
Due: September 24th by 2:30pm
1. Adult males are taller, on average, than adult females. While visiting a recent junior
karate tournament with your little sister, you do not observe an obvious difference in t
14.32 PSET 2
Rachael Meager
September 29, 2015
Due: October 8th by 2:30pm
1. We continue to work with the Hogwarts School of Witchcraft and Wizardry Teacher Ratings data set from Problem Set 1. Estimate a bivariate regression of course evaluations
score o
14.32 PSET 3
Rachael Meager
October 13, 2015
Due: October 22nd by 2:30pm
1. Say whether these statements are True, False, or Uncertain and explain why in each case:
(a) Consider a linear regression with a constant term. Categorical variables, when include
14.32 PSET 4
Rachael Meager
November 3, 2015
Due: November 12th by 2:30pm
1. Your friend Vikram insists that it is a scientific fact that on average married men make more
money, and that this is a causal impact from getting married. He is getting married
14.32: Econometrics
Problem Set 3
due Thursday, October 20, 2016
1. True, False, Uncertain with Explanation:
(a) Dummy variables are variables added to the regression that have no
explanatory power but serve only to increase the number of degrees of
freed
14.32: Econometrics
Problem Set 2
due Thursday, October 13, 2016
1. True, False, Uncertain with Explanation:
(a) The assumption that the errors in a regression model have a normal distribution is necessary to obtain the result that the least squares estim
Serial Correlation
14.32 Spring 2007
Consider the generic model
y t x t t
where all of the normal assumptions hold EXCEPT for random sampling, i.e.,
E i j 0. This violates the Gauss-Markov assumptions, so OLS estimators will
not be BLUE. BUT, they are sti
Implementation of IV and Two-Stage Least Squares
14.32, Spring 2007
Review of IV
In lecture we saw the schooling model
y i S i A i i
where y i is log income of individual i, S i is schooling, and A i is unmeasured ability.
Since ability is unobservable, t
Additional Notes on 2SLS and Simultaneous Equations
14.32, Spring 2007
Recitation 5/11/07
Consider the generic model
qd 0 1p 2x
qs 0 1p 2z
qs qd
In this system, both the supply and demand equations are exactly identified. We
can estimate 1 and 1 using i
More on Differences-in-Differences
14.32 Recitation 4/13/07
In lecture, we went over Card and Kreugers estimating equation:
y it NJ i D 2 NJ i D 2
#
Sometimes its useful to use a 2x2 matrix for thinking about DD. For each of Card
and Kreugers four state-t
Problem Set VI
MIT (14.32)
Spring 2007
A. Wooldridge, problems 15.2, 15.3, 15.7, 15.10
B. Use the data set on the course web site to replicate the OLS and IV estimates of
the returns to schooling for the 1980 Census reported in Table V, Columns 1
through
Cite
Problem Set I
MIT 14.32
Spring 2007
Due: Th, Feb 22
A. From Wooldridge: C.2, C.3, C.4, C.5, C.8, C.10
B. Additional problems
1. This problem asks you to use SAS to conduct a series of sampling experiments.
a. Draw 500 random samples of size 6 from a
Problem Set II
MIT (14.32)
Spring 2007
due: March 8, 2007
A. From Wooldridge: B.10 (more on expectation), 2.4, 2.8, 2.9 (bivariate regression basics)
B. Additional problems
1. Consider E[Y| X] where X is a dummy variable that equals one with probability p
Problem Set IV
MIT (14.32)
Spring 2007
due: April 12
A. From Wooldridge: 5.1, C5.2, 6.4, 7.3, 7.5, C7.10
B. The 14.32 web page contains an ASCII data set (PS4.DAT) with observations on log weekly wages,
log hourly wages, age, sex (1=male), race (1=White,
Problem Set V
MIT (14.32)
Spring 2007
A. From Wooldridge: 8.1, 8.2, 8.5, C8.5, 10.1, C10.5, 12.1, C12.10
B. Additional problems:
This problem set asks you to replicate Kruegers 1993 QJE paper, How Computers
Have Changed the Wage Structure. The paper can b
Econometrics
MIT (14.32)
Spring 2007
Review Problem Set
This problem set is ungraded but may be handed in for 1 bonus point.
A. Using data for the most recent year available from the Bureau of Labor Statistics, compute or do the
following:
1. The joint di
14.32: Econometrics
Problem Set 1
due Tuesday, September 27, 2016
1. Adult males are taller, on average, than adult females. Visiting two recent
American Youth Soccer Organization (AYSO) under-12-years-old (U12) soccer
matches on a Saturday, you do not ob
14.32 PSET 5
Rachael Meager
November 17, 2015
Due: Wednesday 25th November at 5:30pm (submit via email or drop off paper copy with
Tetsuya during office hours, or leave paper copy in Tetsuyas folder on the second floor of E17)
1. True, False, Uncertain wi