Nusrat Farah
Homework-3
Regression Results:
Source
SS
df
MS
Model
Residual
1.0083e+09
18506814.6
3 336100095
231 80116.0803
Total
1.0268e+09
Number of obs
F( 3, 231)
Prob > F
R-squared
Adj R-squared
R
ECN 4973 Introduction to Econometrics
Midterm Exam Review
October 9, 2013
Practice Problems
1.
Consider the example of cigarettes and baby birth weight discussed previo
Solution to the problem 16.14
d) Regression results for Fixed Effect Model:
Fixed-effects (within) regression
Group variable: country1
Number of obs
Number of groups
=
=
60
3
R-sq: within = 0.2764
bet
Solution to the problem 16.14
1)
a) As priori, we expect a negative relationship between compensation and unemployment.
If there is high unemployment then there will be less pressure for wage increase
Random walk with deterministic trend:
If the trend in a time series is a deterministic function of time, such as t and t 2, we call it a
deterministic trend (predictable). If it is not predictable, we
Random walk model: A Random Walk Model (RWM) is a non-stationary process. There are
two types: a) without a drift and b) with a drift. We can perform graphical analysis and
regressions to determine wh
Random walk with drift: Here, Yt = +Yt1 + ut, so we can write Y1 = +Y0 + u1, Y2 = +Y1 +
u2, Y3 = +Y2 + u3 and replacing, Y3 = +Y0 + u1 + u2 + u3 such that Yt = +Y0 +ut.
Therefore, E (Yt) = t+Y0 since
Exercise 11.6:
a) The assumption made by the author is that the error variance is proportional to the
square of Gross National Product (GNP) as described in the postulation. We can justify
this as the
Exercise 10.3:
a)
The numerical value of the intercept and slope coefficients of PGNP and FLR have
changed but their signs are still the same. These variables are still statistically significant. The
ECN 4973 Introduction to Econometrics
Midterm Exam Review
October 9, 2013
Summary of Topics Covered
1.
Basics of Multiple Linear Regression
a. Ordinary Least Squares (