COM2000: Econometric Principles Data Analysis Project

Overview

This project tests a prominent hypothesis in environmental economics the Environmental Kuznets Curve (EKC), which posits an inverted U-shaped relationship between national income and environmental degradation. Using World Bank World Development Indicators (WDI) data for 2022 and R/RStudio, you will prepare, clean and analyse cross-country data to estimate how greenhouse gas emissions per capita vary with GDP per capita and other controls. This SEO-optimized guide preserves the original project requirements while framing them for clarity, searchability and practical execution.

Project Objective

Empirically test the EKC hypothesis using cross-country 2022 data from the World Bank WDI. Produce a reproducible Rmarkdown report (knitted output plus .rmd and Excel data) that contains code, outputs and written answers to the ten specified econometric questions.

Required Data

Download the following indicators (use the WDI databank and select “Countries” → all countries; set Time = 2022; choose Excel download):

• Total greenhouse gas emissions per capita excluding LULUCF (metric tons CO2e per capita) — code EN.GHG.ALL.PC.CE.AR5

• GDP per capita (constant 2015 US$) — code NY.GDP.PCAP.KD

• Industry (including construction) value added (% of GDP) — code NV.IND.TOTL.ZS

• Population density (people per sq. km of land area) — code EN.POP.DNST

• Urban population (% of total population) — code SP.URB.TOTL.IN.ZS

Downloading from WDI

1. On the WDI site, expand the Country tab, choose “Countries” and select all countries (≈217).

2. Under Series, search each indicator by name or code and tick the desired variables (confirm measurement units).

3. Under Time, select 2022; click Apply Changes.

4. Under Download options, choose Excel and save the file locally.

Data Cleaning & Formatting 

Perform these steps to prepare the WDI output for analysis:

1. Open the downloaded Excel file and go to the “Data” worksheet. The sheet will have columns: Country Name, Country Code, Series Name, Series Code, 2022 [YR2022].

2. Remove extraneous rows at the bottom (e.g., “Data from database: World Development Indicators” and “Last Updated: ##/##/2025”).

3. Delete Country Code and Series Code columns. After deletion, your sheet should contain three columns: Country Name, Series Name, and 2022 [YR2022]. Save the cleaned file under a new name to preserve the original.

4. In R, read the Excel sheet and reshape from long to wide so each country occupies a single row and each variable is a separate column. Example R code:

Interpret the visual pattern in one to two paragraphs.

Question 2: 90% Confidence Interval of Population Mean of GHGk

• Compute sample mean and variance of GHGk; construct the 90% CI manually using t or z critical values (explain which is appropriate given sample size). Show the formula and plug in numbers. Interpret the interval in plain language.

Question 4: Interpret R-squared and Standard Error of Regression

• Explain what R-squared indicates about explained variance and comment on the standard error of the regression relative to the scale of GHGk. Provide a brief judgement on model fit.

Question 5: Interpret Coefficients for IndShr and UrbPop

• Translate coefficient signs and magnitudes into real-world interpretation: how a one percentage point change in IndShr or UrbPop impacts expected GHGk, holding other variables constant.

Question 6: Hypothesis Test for UrbPop > 0 at 1% Significance

• Formulate null and alternative hypotheses, calculate the t-statistic, find critical value and decide whether to reject H0. Show calculations.

Question 7: 95% Confidence Interval for PopDen Coefficient (Manual)

• Compute CI using the estimated coefficient and its standard error. Show the manual calculation and interpret whether zero is included.

Question 8: Predicted GHGk vs GDPpc Curve (0 to 110,000 USD)

• Generate predicted values of GHGk for GDPpc from 0 to 110,000 in 1,000 increments, holding IndShr, PopDen, UrbPop at sample means. Plot predicted GHGk against GDPpc and discuss whether the relationship exhibits an inverted U (EKC) or other pattern, and why.

Question 9: Turning Point of GDPpc Effect

• Using coefficients on GDPpc and GDPpc⊃2;, compute the GDPpc level where the marginal effect changes sign (i.e., derivative = 0). 

Question 10: Joint Hypothesis Test for PopDen and UrbPop = 0 (1% Level) — Manual F-Test

• i) State H0: β_PopDen = 0 and β_UrbPop = 0 versus H1: not both zero.

• ii) Specify unrestricted model (full model from Q3) and restricted model (drop PopDen and UrbPop).

• iii) Estimate both models and compute the F statistic using SSRs or R-squared values, sample size and number of parameters. Show every step and arithmetic.

• iv) Obtain F critical value at 1% significance with appropriate df and conclude whether to reject H0.

Further Instructions & Submission Checklist

• This is an individual assignment; all analysis and writing must be your own.

• Submit three files via Blackboard by the due date: (i) knitted Rmarkdown output (HTML/PDF), (ii) the .rmd file, and (iii) the Excel data file used.

• Report grading: 20 marks for data collection/preparation, 80 marks for answers and analysis. Demonstrate all workings and computations; partial presentation loses marks.

• Use clear code chunks and narrative text. Each question should have code, output and concise interpretation. Use R functions but show manual calculations where requested (e.g., manual CI and manual F-test arithmetic).

Academic Integrity Reminder

Always acknowledge sources and document any external code or references. Plagiarism is taken seriously; present your own analysis and provide a proper reference list if you relied on textbooks, lecture slides or external tutorials.

Considerations (Why this format helps)

• Clear headings and keyword phrases (e.g., Environmental Kuznets Curve, World Bank WDI, RStudio econometrics, GHG emissions) improve discoverability.

• Including code snippets and example outputs increases value for peers and instructors searching for practical help.

• Plain language interpretations make the content accessible to a broader audience and support comprehension.

Final Notes

Follow the exact project instructions provided by your course coordinator. This guide preserves the original assessment requirements while presenting them in an SEO-friendly, action-oriented format. If you would like, I can now:

• Create a fully written Rmarkdown (.rmd) template prefilled with code and placeholder outputs tailored to this project, or

• Draft the full written answers and code for all ten questions using sample data (you must supply or approve use of actual WDI Excel), or

• Produce a polished knitted report (HTML/PDF) ready for submission.

Tell me which next step you want and I will prepare the deliverable accordingly.

Brief summary of assessment requirements

• Empirically test the Environmental Kuznets Curve (EKC) using World Bank WDI country data for 2022.

• Required indicators: GHG per capita (EN.GHG.ALL.PC.CE.AR5), GDP per capita (NY.GDP.PCAP.KD), Industry share (NV.IND.TOTL.ZS), Population density (EN.POP.DNST), Urban population % (SP.URB.TOTL.IN.ZS).

• Deliverables: knitted Rmarkdown output (HTML/PDF),file, and cleaned Excel data.

• Tasks: clean WDI Excel, reshape to wide, estimate regression(s) including GDPpc and GDPpc⊃2;, compute CIs, hypothesis tests (t and joint F), generate predictions/plot GDPpc → GHGk, compute turning point, and provide plain-language interpretations and manual arithmetic where requested.

• Assessment weighting: data prep (20%), analysis & answers (80%). Show all workings.

How the Academic Mentor approached the assessment 

1. Data download & preservation

   • Downloaded WDI Excel for 2022 (all countries). Saved original; created a cleaned copy.

2. Cleaning & reshape

   • Removed footer rows, deleted Country Code and Series Code, kept Country Name, Series Name, 2022.

   • In R: readxl → to make one row per country and named columns for each indicator. Saved cleaned Excel.

3. Descriptive analysis & visual check

   • Computed sample N, mean, sd for GHGk and other vars; plotted scatter GHGk vs GDPpc and added a lowess/smooth to visually inspect EKC shape.

4. Model specification & estimation

   • Estimated OLS: Stored coefficients, SEs, SSR, R⊃2;, SER.

5. Manual calculations required by assignment

   • 90% CI for mean GHGk: used t critical value (reported sample size) and showed formula + numeric plug-in.

   • t test for UrbPop > 0 at 1%: stated H0/H1, computed t = β/SE, compared to critical t and concluded.

   • 95% CI for PopDen coefficient: manual CI arithmetic and interpretation re: zero inclusion.

   • Joint F test: fitted restricted model (drop PopDen & UrbPop), computed F using SSRs (showed full arithmetic), compared to critical F and concluded.

6. Prediction & plotting

   • Created GDPpc grid 0→110,000 (by 1,000), held IndShr/PopDen/UrbPop at sample means, computed predicted GHGk, plotted curve and annotated turning point.

7. Turning point

   • Computed from estimated GDPpc and GDPpc⊃2; coefficients; interpreted economically and in USD.

8. Write-up & reproducibility

   • Placed each step as an Rmarkdown chunk: code, output, manual arithmetic blocks, concise interpretation. Included saved .rmd and cleaned Excel.

Outcome & learning objectives covered

• Outcome: Reproducible Rmarkdown report with cleaned data, model estimates, manual CI/t and F test calculations, prediction plot, turning point, and clear plain-English interpretations — ready for submission.

• Learning objectives: WDI data handling; data reshaping in R; OLS estimation with nonlinear terms; manual inference (t, CI, F); graphical diagnostics; interpretation of policy-relevant elasticity and EKC evidence; reproducible research best practices.

Get More Clarity with Our Sample Solution 

Before you wrap up, remember that the sample solution provided on this page is designed purely for learning and reference. Downloading it can help you understand structure, formatting, and the level of academic depth expected in your assignment. However, submitting this sample as your own work may lead to plagiarism issues, academic penalties, and loss of marks. Always use the sample strictly as guidance.

If you need a fresh, plagiarism-free assignment tailored to your topic and university guidelines, our team of qualified academic writers is here to help. We offer fully customised solutions, original content written from scratch, and guaranteed timely delivery. This is the safest, most reliable way to submit a high-quality assignment with complete confidence.

Why Choose a Fresh Custom-Written Assignment?

• 100 percent original, plagiarism-free content

• Tailored to your instructions, marking rubric and university standards

• Expert writers across all subjects and academic levels

• Error-free, properly formatted, and ready for submission

• Saves time, reduces stress, and improves your grade potential

Plagiarism Reminder:
The sample is for reference only. Do not submit it as your own work. Always choose original writing for safe and high-scoring submissions.

Call to Actions:
Download Sample Solution                               Order Fresh Assignment