Scribd
Upload
4 views

Assignment Regression 6.5.2025

The document outlines the coursework requirements for EM4810E Business Forecasting at Hanoi University of Science and Technology, due on May 18, 2025. It includes instructions for data cleaning, analysis through scatter plots and regressions, and a final report with a word limit of 400 words discussing convergence in output and regression comparisons. The coursework is worth 20% of the total mark and involves analyzing economic data from 1960 to 2019 for 110 countries.

Uploaded by

Hoàng Nguyễn Việt
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
4 views

Assignment Regression 6.5.2025

The document outlines the coursework requirements for EM4810E Business Forecasting at Hanoi University of Science and Technology, due on May 18, 2025. It includes instructions for data cleaning, analysis through scatter plots and regressions, and a final report with a word limit of 400 words discussing convergence in output and regression comparisons. The coursework is worth 20% of the total mark and involves analyzing economic data from 1960 to 2019 for 110 countries.

Uploaded by

Hoàng Nguyễn Việt
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/ 3

School of Economics and Management,

Hanoi University of Science and Technology.

EM4810E. Business Forecasting - Coursework


Instructor: Ha Thi Thu Trang (PhD)

Due: 18 May 2025, 4PM

Presentation Date: 20 May 2025, 7AM

General Instructions
Please prepare your answers in one Word document with Rscripts in Appdendix and then
save as a PDF file before uploading them to MS Team. Submit only the final document with
graphs, regressions, and answers to the questions. Answers should be succinct and within
word limits. The coursework is worth 20% of the total mark.

Download data from MS Team

Variables: year, countrycode, rgdpe, pop, and csh_i for 1960-2019.

Data Cleaning
Save the following variables: year, countrycode, rgdpe, pop, and csh i for all the countries in
the dataset for the years 1960 to 2019.

Remove from the dataset of 183 countries, all the countries for which we do not have all the
observations for rgdpe, pop, csh i for the years 1960 to 2019. Also eliminate the
observations for Venezuela as it is an extreme outlier (according to the dataset real per
capita GDP dropped from just over 19,000 to 250 between 2012 and 2019)

After cleaning, data should include 110 specific countries.

Part I
1. Draw the scatter plot with per-capita real GDP in 1960 (relative to the USA) vs.
average per-capita growth rate from 1960 to 2019.
[10 marks]
2. Run two regressions:
(1) gi = α + β ln(yi) + εi
(2) gi = α + β1 ln(yi) + β2 ln(ni + d) + β3 ln(si) + εi

where the i index represent each country, yi is per-capita real GDP (output)
(rgdpet/popt) for the initial period in the regression (i.e., t=1960), g i is the average of the
growth rates of per-capita real GDP (yt – yt-1)/ yt-1 for the entire period (i.e., how the
economy grew on average, from the initial year 1960 to the final year 2019), n i is the
average growth rate of the population (popt – popt-1)/ popt-1 for the entire period (i.e.,
1960 to 2019) and d is the rate of depreciation (which we will assume is identical across
all countries and equal to 0.05, so you need to add this number to the values of ni before
you run your regressions), and si is the average of the variable csh i for the entire period
(i.e., 1960 to 2019). [10 marks]

Part II
Complete the following for two periods:

Group 1 + 5:

a) Create a scatter plot with the average annual growth rates for each country in the
sample from 1960 to 1995 (vertical axis) against per-capita real GDP in 1960 (relative
to the USA in 1960). [10 marks]
b) Estimate the regressions (1) and (2) for this time period and show the full results.
[10 marks]
c) Create a scatter plot with the average annual growth rates for each country in the
sample from 1996 to 2019 (vertical axis) against per-capita real GDP in 1996 (relative
to the USA in 1996). [10 marks]
d) Estimate the regressions (1) and (2) for this time period and show the full results.
[10 marks]
Group 2 + 4:
a) Create a scatter plot with the average annual growth rates for each country in the
sample from 1960 to 1997 (vertical axis) against per-capita real GDP in 1960 (relative
to the USA in 1960). [10 marks]
b) Estimate the regressions (1) and (2) for this time period and show the full results.
[10 marks]
c) Create a scatter plot with the average annual growth rates for each country in the
sample from 1998 to 2019 (vertical axis) against per-capita real GDP in 1998 (relative
to the USA in 1998). [10 marks]
d) Estimate the regressions (1) and (2) for this time period and show the full results.
[10 marks]
Group 3+6+7:
a) Create a scatter plot with the average annual growth rates for each country in the
sample from 1960 to 1998 (vertical axis) against per-capita real GDP in 1960 (relative
to the USA in 1960). [10 marks]
b) Estimate the regressions (1) and (2) for this time period and show the full results.
[10 marks]
c) Create a scatter plot with the average annual growth rates for each country in the
sample from 1999 to 2019 (vertical axis) against per-capita real GDP in 1999 (relative
to the USA in 1999). [10 marks]
d) Estimate the regressions (1) and (2) for this time period and show the full results.
[10 marks]
Part III
Word limit: 400 words total for all two questions.
1. Given your results in Part I and II, what do we learn about convergence in output
over time since 1960? Hint: Focus on the estimates of (1). Explain. [20 marks]
2. Comparing the regressions (1) and (2) for both Parts I and II, which coefficient on
per-capita real GDP is larger (in absolute value) and more statistically significant?
Why? [20 marks]

---------------------------------------------------------------------------------------------------

You might also like