Econ471: Intro Applied Econometrics
Walter Sosa-Escudero
First Midterm
RULES: Read the exam very carefully and answer all questions. The exam is worth 100 points. The
rst two questions are worth 25 points each, and the last question is worth 50 points. Yo
The Unbiasedness Property
Model Specication
Model Specication: Precision and Bias
Walter Sosa-Escudero
Econ 507. Econometric Analysis. Spring 2009
April 1, 2009
Walter Sosa-Escudero
Model Specication: Precision and Bias
The Unbiasedness Property
Model Spe
Random Regressors
Walter Sosa-Escudero
Econ 471. Econometric Analysis. Spring 2009
April 11, 2012
Walter Sosa-Escudero
Random Regressors
The Classical Linear Model:
1
Linearity: Y = X + u.
2
E (u) = 0
3
No Multicollinearity: (X ) = K . X not-random
4
No h
HW#2
1. (a) The p- value means that the probability of =0 is 0.1262. It indicates a small
probability that x has little effect on y.
Graph: As all is under t-distribution:
It means 12.62% of the estimated will fall in the shaded area.
(b) According to the
Final Remarks on
Econometrics
Walter Sosa-Escudero
wsosa@illinois.edu
On the scope of econometrics
Vertical integration of statistics: the part of
statistics produces within economics due
to its salient specificities: dependencies
(autocorrelation, dynam
Introduction
Multicollinearity and Micronumerosity
Multicollinearity
Walter Sosa-Escudero
Econ 507. Econometric Analysis. Spring 2009
March 10, 2009
Walter Sosa-Escudero
Multicollinearity
Introduction
Multicollinearity and Micronumerosity
The Classical Li
Measurement Errors
Walter Sosa-Escudero
Econ 471. Econometric Analysis. Spring 2009
April 21, 2009
Walter Sosa-Escudero
Measurement Errors
Motivation
One particular case of explained variable that is random arises
when it is measured with errors.
Walter S
Autocorrelation
Walter Sosa-Escudero
Econ 471. Econometric Analysis. Spring 2009
April 23, 2009
Walter Sosa-Escudero
Autocorrelation
Time-Series Observations
Consider the following model
Yt = 1 + 2 X2t + + K XKt + ut ,
, t = 1, 2, . . . , T
Here t denotes
Dynamic Regression
Walter Sosa-Escudero
Econ 471. Econometric Analysis. Spring 2009
April 28, 2009
Walter Sosa-Escudero
Dynamic Regression
Motivation
Consider the following dynamic model
Yt = 1 + 2 Xt + Yt1 + ut ,
where t = 1, . . . , T indicates time-ser
Applied Econometrics. Walter Sosa Escudero
Heteroscedasticity
a) Tests for heteroscedasticity
Earnings function (Johnston and DiNardo p. 172)
1. Initial OLS estimates
LS / Dependent Variable is LNWAGE
Date: 10/04/97 Time: 15:40
Sample: 1 100
Included obse
University GPA
Tong Wu
Bolong Zhang
Lu Liu
Data: From 105 Computer science major students
Explanatory variable: high school gpa, math SAT score, Verbal S
AT score, computer science GPA
.
We did theExplained variable: variables together. Since SA
regressio
Econ471: Introduction to Applied Econometrics
Walter Sosa-Escudero
First Midterm
RULES: Read the exam very carefully and answer all questions. The exam is worth 90. Each
question is worth 30 points. You can use a sheet of notes and a calculator. You have
Two small things
Walter Sosa-Escudero
Econ 471. Applied Econometrics. Spring 2009
March 5, 2012
Walter Sosa-Escudero
Two small things
Linear vs. Quadratic
Union eect problem: in spite of being statistically signicant, is
the quadratic specication economic