Scribd
Upload
6 views4 pages

Review Problems

The document is a practice sheet for ECON 3081, containing questions on econometric concepts such as OLS, error terms, model comparison, and productivity across sectors. It includes true/false statements, calculations for coefficients and errors, and interpretations of regression results. The document also addresses structural change testing and regression modeling based on age and consumption.

Uploaded by

Mostafa Allam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
6 views4 pages

Review Problems

The document is a practice sheet for ECON 3081, containing questions on econometric concepts such as OLS, error terms, model comparison, and productivity across sectors. It includes true/false statements, calculations for coefficients and errors, and interpretations of regression results. The document also addresses structural change testing and regression modeling based on age and consumption.

Uploaded by

Mostafa Allam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

ECON 3081 (02)

Practice sheet
Fall 2024

Question One:

State if the following statements are true or false and justify your answer.

1. The variance of the error term should be constant


2. Autocorrelation refers to the trend in the variables of interest
3. The correlation between the independent variables does not affect the precision of the
estimates
4. The error term should be normally distributed
5. OLS methodology always works
6. No tests can compare different models
7. Endogeneity refers to the inability to guarantee a causal effect of the variable of interest
8. Omitted variables do not cause bias
9. The error term and the independent variable shall be correlated
10. Adjusted R squared accounts for number of observations and number of variables

Question Two

Based on OLS in matrix form and given the following observations on X and Y:

X: 1, 2, -1, 0, 4
Y: 4, 3, 5, -2, 1

1. Compute the vector of coefficients.


2. Compute the vector of errors
3. Compute the coefficients’ variance-covariance matrix
4. Discuss the significance of coefficients at the 5% level if tabulated t is 2.5.

Question Three:

0.0099941 ?
𝑉"𝛽$% = ( 1
? 0.000007547

Assuming that V(𝛼3 + 𝛽$ ) = 0.00926997, Calculate the missing parts in the matrix.

V(alpha + beta) = var(alpha) + var(beta) + 2 cov(alpha, beta)

Cov = (0.00926997 - 0.0099941 − 0.000007547) / 2

1
Question Four:

𝑌!∗ = α + β 𝑇! + ε

𝑌!∗ 𝑖𝑠 𝑑𝑒𝑓𝑖𝑛𝑒𝑑 𝑎𝑠 (𝑌! − 𝑀! )


𝑌! is the maximum number of sales
𝑀! is the current level of sales
𝑇! is time in weeks

The model is estimated as follows:

𝑌F∗= −10.4 + 0.3 𝑇!

1. Define 𝑌!∗
It is the gap between potential and current sales
2. What is the interpretation of 𝛽$
Each 1-week results in a 0.3 higher difference in potential and current sales
3. At what time will the sales reach the maximum?
At Y* = zero
T = 10.4/0.3 = 34.66 weeks
Sales will reach the maximum value after 34.66 weeks

Question Five:

Assuming the economy is composed of three sectors: Agriculture, Manufacturing, and Services
indexed by I

𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑣𝑖𝑡𝑦 = 𝛼 + 𝛽# 𝐼# + 𝛽$ 𝐼$ + 𝜀
I1 = 1 if Agriculture is observed
0 otherwise
I2 = 1 if Manufacturing is observed
0 otherwise

S
𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝚤𝑣𝚤𝑡𝑦 = 3733 – 2300 I1 – 3166 I2

1. What is the average productivity for each sector?


Av productivity of services = 3733
Av productivity agriculture = 3733 – 2300 = 1433
Av productivity of manufacturing = 3733 – 3166 = 567
2. Interpret 𝛽#
On average, agriculture has lower productivity than services by 2300
3. If the econometrician defines manufacturing as the reference sector, what will be the
constant?
567

2
Question Six:

Given the estimated model for 18 observations:

Y = 2 + 0.5 X with ESS = 3000

We divide the data and estimate two models:

1- Y = 0.3 + 2 X where ESS = 1000 for the first 9 observations


2- Y = 9 + 0.01 X where ESS = 1100 for the other 9 observations

Test for structural change if F tabulated is 11.6.

Answer:

Ho: there is no evidence for structural change


H1: there is structural change

F Chow = ((ESSr – ESSur)/m) / ESSur/(n1 + n2 – 2k)

= ((3000 – 1000 – 1100) / 2) / (1000+1100)/ (18 – 4)


= 450 / 150 = 3

Since Fchow < F tabulated


Then we do not reject H0 that we do not observe evidence of structural change and we reject H1

Question Seven

For a sample size of 40 observations, we regress two models:

Model 1: 𝑌 = 𝛼 + 𝛽# 𝐿 + 𝛽$ 𝐾 + 𝜀 , SSE = 1,100


Model 2: 𝑌 = 𝛼 + 𝛽# 𝐿 + 𝛽$ 𝐾 + 𝛽% 𝐾 $ + 𝛽& 𝐿$ + 𝜀 , SSE = 1,000

Which model should the econometrician employ?

Question Eight

Assuming a regression model where Consumption Ct is determined by Age Gt

1- Define a regression model where consumption is determined only by Age.


2- Assume a Piecewise regression where 30 years is the threshold age point. Re-define the
regression model.
3- What is the expected consumption before and after the threshold age?

3
Question Nine

Linearize the following model and clarify if OLS can estimate it using the simple form and
deduce the parameters.

,#
𝑒 '/) = 𝑒 *+ ) +-.+ 0

I = 1 if I = i* and 0 otherwise

You might also like