1. Countries have a keen interest in exploring the drivers of their
sectoral energy consumption, including INDUSTRIAL energy use. Please examine the log
final energy use by industry
“ln_indpc”
a) Design a regression model to predict “lnind_
pc”.
Choose the explanatory variables to include, and whether to include them as dummies/ logs/ polynomials/ interactions as you feel appropriate.
Present at least two final models.
(1)
One with a linear per capita term (or its logs) [Model 1],
(2)
one with a quadratic term (or its log) [Model 2].
Which model do you think is more appropriate? How do you explain the quadratic model?
Present the
results of the descriptive statistics and your final regression models in tables. You should have 3 Tables to gain full points.
(3 marks)
Discuss the logic behind the specification for
BOTH of your models
-why you used logs or levels
(1 mark)
-how you interpret polynomials
(2 marks)
-why and how you used dummies and how to interpret these
(1 mark)
Discuss the statistical significance
of the explanatory variables in your models.
(2 marks)
b) Discuss how you have designed your model with
reference to the “Gauss Markov”
assumptions and whether these assumptions are likely to be met.
(2 marks)
Interpret
the results of
THREE
of your explanatory variables including income per capita, which you consider to be the
key drivers
of
per capita industrial energy consumption.
(3 marks)
(Total: 14 marks) (550 words, 3 tables, 4 calculations)
There will be up to 2 additional marks awarded for presentation of your answers (neat formatting of tables and clear expression of answers in full sentences).