Scribd
Upload
34 views3 pages

Assignment Submission Guidelines and Questions

The document provides instructions for submitting an assignment. It must be submitted via email within two days of the final exam before 10 pm as a PDF file. The subject line and file name should include the student's name and section. The assignment contains questions about regression analysis, using data files to answer questions, and examining determinants of house prices and CEO salaries.

Uploaded by

Yaregal Bante
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
34 views3 pages

Assignment Submission Guidelines and Questions

The document provides instructions for submitting an assignment. It must be submitted via email within two days of the final exam before 10 pm as a PDF file. The subject line and file name should include the student's name and section. The assignment contains questions about regression analysis, using data files to answer questions, and examining determinants of house prices and CEO salaries.

Uploaded by

Yaregal Bante
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

Assignment

Instruction on Submission
1. The assignment must be submitted via email within two days after the final exam
before 10:00 pm
2. The recipient email is: [Link]@[Link]
3. The submission must have cover page which contains full name, ID No. and Section
4. The softcopy must be in PDF format
5. In the submission email, the subject is “Your Name-Section” and the file name
should also be “Your Name”
6. The objective of the assignment is to test your knowledge not how good you are at
copying and pasting from the internet or friends. Your work will not be marked if
found to be copied from books or internet or friends.

Questions
1. What is the difference between regression and correlation?
2. What is ordinary least square?
3. What is the role of the error term ui in regression analysis?
4. What is the difference between the error term and the residual?
5. What are the assumptions of ordinary least square? Explain them?
6. Explain the nature, cause, and consequence of heteroscedasticity?
7. Explain the nature, cause, and consequence of multicollinearity?

8. Use the data in EDUC (see the other file in the assignment folder) for this exercise.
i. What are the lowest and highest years of education?
ii. What is the average education level in the sample?
iii. What is the average hourly wage in the sample? Does it seem high or low?
iv. How many women are in the sample? How many men?
v. How many people have experience above the mean?
vi. Estimate the equation:
wage = 0+1educ+2exper+3tenure +u
a) Interpret each parameter
Page 1 of 3
b) Repeat part (a), but with log(wage) as the dependent variable (Search from Google the
Stata command to transform a variable to log form).

9. The file MGT (see the other file in the assignment folder) contains data on chief executive
officers and can be used to examine the effects of firm performance on chief executive
officer salary. Use this STATA file to answer the following questions.
i. Estimate a model relating annual salary to firm sales, profits, and market value.
ii. Interpret each coefficient? Which variables are statistically significant and at what level?
iii. Would you say that these firm performance variables explain most of the variation in
CEO salaries? Why do you say that?
iv. Add the variable ceoten to the model in part (i). What is the estimated percentage return
for another year of CEO tenure, holding other factors fixed?
v. Create natural logarithmic forms of salary, sales, and market value variables. Why can
profit not be included in logarithmic form?
vi. Estimate a model relating annual salary (in log form) to firm sales (in log form), profits,
and market value (log form). How do you interpret the coefficient of log (firm sales)?
Which variables are statistically significant and at what level?
vii. Estimate a model relating annual salary to firm sales (in log form), profits, and market
value (log form). How do you interpret the coefficient of log (firm sales)?
viii. Estimate a model relating annual salary (in log form) to firm sales, profits, and market
value (log form). How do you interpret the coefficient of firm sales?
ix. Suppose you develop the following quadratic model:
salary = 0+1sales+2age+3age2 +u
a) What is the effect of age on annual salary?
b) What is the turning point? What does it mean?

Page 2 of 3
10. You want to examine the determinants of house price in Addis Ababa. You collect data and
get the following equation.
price = -19.315 + .128 sqrmeter + 15.198 bdrms
n =540, R2 =0.36
where price is the house price measured in thousands of birrs, sqrmeter is the square meter of
the house, and bdrms number of bed rooms.
a) How do you interpret the intercept?
b) How do you interpret the R2?
c) What is the estimated increase in price for a house with one more bedroom, holding
square meter constant?
d) What is the estimated increase in price for a house with an additional bedroom that is
140 square meter in size?
e) The first house in the sample has sqrmeter = 2,438 and bdrms = 4.
i) Find the predicted selling price for this house from the OLS regression line.
ii) The actual selling price of the first house in the sample was Birr 300,000 (so
price=300). What is the residual for this house?

Page 3 of 3

Common questions

Powered by AI

An R-squared value of 0.36 implies that approximately 36% of the variance in house prices can be explained by the model variables—square meter size and number of bedrooms—while 64% is unexplained, possibly due to omitted variables or noise .

Logarithmic transformation is used to linearize relationships, stabilize variance, and make interpretation relative, representing elasticities (percentage changes). It simplifies multiplicative relationships, making coefficient interpretation as percentage changes rather than absolute ones .

To determine if firm performance variables explain variations in CEO salaries, analyze the R-squared value for the proportion of explained variance, conduct hypothesis tests on model coefficients to assess significance levels, and evaluate adjusted R-squared for model fit improvement as additional variables are included .

In the given regression model, the coefficient for 'sqrmeter' is 0.128, meaning that for each additional square meter, holding all else constant, the house price increases by approximately 128 birr .

Multicollinearity can inflate the variances of parameter estimates, leading to large standard errors of the coefficients, which may result in statistically insignificant predictors even when they are significant. It complicates the interpretation of the model and decreases the precision of the parameter estimates .

The predicted triangle inequality involves examining sums of squares in a regression analysis (e.g., Total SS = Explained SS + Residual SS). Violations suggest potential issues with linearity or model specification, indicating the need for model refinement or variable transformation .

Adding the variable ceoten (CEO tenure) to the salary regression model helps control for the effect of tenure on salary, potentially increasing the model's explanatory power. It allows for estimating the percentage return of another year of CEO tenure, revealing dynamic salary structure over time .

Heteroscedasticity in a regression model is identified through tests like the Breusch-Pagan or White test, and visualization techniques such as plotting residuals against fitted values. Its presence indicates that the variance of errors is not constant, which can lead to inefficiency and unreliable statistical inference of OLS estimates .

The assumptions of OLS, such as linearity, independence, homoscedasticity, and normality of residuals, directly affect the model's interpretability and validity. Violations can lead to biased estimates, inefficiency, or inconsistency in the parameter estimates, making inferences based on the model potentially misleading .

Failure to follow the assignment submission guidelines could lead to the submission being disqualified or not marked at all. This includes submitting after the deadline, not using the specified email or format, or presenting an assignment that appears to be copied from others. Such actions could demonstrate a lack of adherence to the instructions designed to test individual knowledge and understanding .

You might also like