ECON1272 Basic EconometricsR

Java Python ECON1272

Basic Econometrics

Individual Assignment

This is an individual assignment where you must work alone.  You must submit an electronic copy of your assignment in Canvas in pdf, doc or docx format and your R code must be copied into Question 4.  Hard copies will not be accepted. Show your calculations (if any) as well as answering the questions in clear full sentences.  Log referrers to natural logarithm!

Use the dataset: WDI_2404.RData

Greenhouse gas emissions (GHG) per capita is an important contributor to global warming. The Paris Agreement (2015) foresees participating states to submit a Nationally Determined Contribution (NDC) to mitigate greenhouse gas emissions. You are a newly hired analyst tasked to model greenhouse gas emissions per capita worldwide. Assume that the outgoing research officer had started working on the econometric model to assess some of the drivers of GHG emissions. Now as an incoming research officer your job is to finish this research. Your variables of interest are:

TGHG =         Total greenhouse gas emissions (kt of CO2 equivalent)

Pop     =         Population, total

GDPpc=          GDP per capita (constant 2015 US$)

Mnf    =         Manufacturing, value added (% of GDP)

REC    =          Renewable energy consumption (% of total final energy consumption)

RIF     =          Renewable internal freshwater resources per capita (cubic meters)

TNRR =         Total natural resources rents (% of GDP)

Trade   =         Trade as % of GDP

LPI      =         Livestock production index

Dependent variable:

Greenhouse gas emissions per capita (GHGpc): We would like to estimate the relationship of other factors with this variable. It is defined as a fraction of Total greenhouse gas emissions over Total Population.

Hint: Please create the level form. of your dependent variable in R, greenhouse gas emission per capita for each country first.

Explanatory variables:

GDP per capita (GDPpc): The richer a country is, some scholars expect higher greenhouse gas emissions per capita. Whether or not this relationship is linear or even reversible at high GDP per capita levels is hotly debated (see the concept of the Environmental Kuznets Curve).

Manufacturing, value added as % of GDP (Mnf): A larger manufacturing sector of a country, is likely associated with more emissions.

Renewable energy consumption in % of total final energy consumption (REC): When a country has a larger share of renewable energy consumption, we expect it to be associated with a lower cp. emissions of greenhouse gas per capita.

Renewable  internal dai 写ECON1272 Basic EconometricsR freshwater  resources  per  capita  (RIF):  Freshwater  reserves  are  an indicator of hydro-electric potential, and thus a cleaner generation portfolio.

Total natural resources rents in % of GDP (TNRR): High levels of natural resources are often associated with the concept of “resource lock-in”, or “fossil-fuel heavy pathways” . The higher the natural resource endowment and rent, according to this theory, the higher their expected usage in the economy.

Trade as % of GDP (trade): Countries with higher levels of trade theoretically have better access to newer and cleaner technologies.

Livestock production  index:  Livestock production index includes meat and milk from all sources, dairy products such as cheese, and eggs, honey, raw silk, wool, and hides and skins (2014-2016 =  100). Higher levels  of livestock are often associated with higher emissions (especially methane).

All data originate from the World Bank (WDI). Please assess whether the above variables are truly associated with GHG emissions, and if yes, how. Answer the following questions:

QUESTIONS:

1)        Use R to create your dependent variable first, then run the following cross-sectional regression. (Please note the natural logs and construct these in R as needed):

log(GHGPc) = β0  + β1 log(GDPPc) + β2 REC + β3 MNF + β4 log(RIF) + β5 TNRR + β6 LPI + u

(Equation 1)

a.    Present your regression results in a table below (R output):   4 marks

b.    Interpret the constant (2.5 marks) and its p-value (1.5 marks).    4 marks

c.    Interpret the coefficient on GDP per capita and its p-value (1.5 marks each).    3 marks

d.    Interpret the coefficient on manufacturing value added and its p-value (1.5 marks each).    3 marks

e.    Interpret  the   coefficient  on  renewable   energy  consumption   (as   %  of   total  energy consumption) and its p-value (1.5 marks each).    3 marks

f.    Interpret the coefficient on the livestock production index and calculate its t-stat.

Interpret the calculated t-statistic (1.5 marks each).    3 marks

g.    Interpret the R2 of the regression.           2 marks

h.    Explain heteroscedasticity, its consequences (2 marks) and present the results of Equation

(1) with heteroscedasticity robust standard errors (3 marks).   Explain  if any  of your coefficient significance levels change (1 mark).          6 marks

2)  Describe each of the Gauss-Markov assumptions and specify if they are likely to hold for the regression in Question 1 or not.           10 marks

3)   Present a functioning R code reproducing the results. This is a critical part of the assignment

without which we’ll initiate a plagiarism check