Transcript US Oil Price Differential

DSCI 5180: Introduction to the Business
Decision Process
Case Study 2
Constructing a Demand Curve for
Crude Oil
© 2013 Nick Evangelopoulos, ITDS Dept., Univ. North Texas
slide 1
Review of Microeconomics/Price Theory
DSCI 5180
Decision Making
 A Demand Curve shows
the relationship between
price and consumption
 Plots Price on the vertical
axis and Consumption
on the horizontal axis
slide 2
Using a Demand Curve
DSCI 5180
Decision Making
 A demand curve can be used in
business planning
 For example, if a certain accident
(spillage, explosion, etc.) results
in a temporary reduction of the
total quantity of an essential raw
material offered for sale in the
market, the demand curve can
help you estimate the expected
price increase so that supply
and demand are stabilized
 Knowing the expected price
increase allows you to adjust your
budget
Price
B
Equilibrium
moves from
A to B
A
Quantity
consumed
slide 3
Oil consumption data
DSCI 5180
Decision Making
 File USOilDemand1970-2002.xls contains data related to demand for
crude oil in the United States in the 1970-2002 period. Data were
obtained from the U.S. Dept. of Energy and U.S. Department of Labor
Web sites.
Y
AdjOilPrice
X1
USPop
X2
USOilCons
X3
WorldOilCons
Inflation Adjusted U.S. Crude Oil
Price, base year = 2005 (in $)
Total Midyear Resident U.S.
Population
U.S. Petroleum Consumption in
million barrels per day
World Petroleum Consumption in
million barrels per day
slide 4
Not a clean demand curve
DSCI 5180
Decision Making
 A preliminary plot of AdjOilPrice versus USOilCons does
not provide a clean demand curve!
slide 5
Why not a clean demand curve?
DSCI 5180
Decision Making
 This happens because our data spans a number of
years during which many things changed, including
population and oil consumption habits and needs
 The price/quantity equilibrium points need to be
adjusted so that they correspond to a single demand
curve.
Price
Price
Quantity
Quantity
slide 6
Drivers of US Oil Consumption other than Oil Price
DSCI 5180
Decision Making
If Oil Prices in the US were held constant, US Oil Consumption would be
driven by such factors as:
•Oil Availability (World Oil Production)
•Population Growth (US Population)
•Spending Habits of the US Consumers (Total US Consumption)
Based on these drivers, we fit a regression model that explains US Oil
Consumption. The unexplained part (residuals) is a US Oil Consumption
Differential.
US Oil Consumption = f(World Oil Production, US Population, US
Consumption)
+ US Oil Consumption Differential
slide 7
Drivers of US Oil Consumption other than Oil Price
DSCI 5180
Decision Making
The regression model has a good fit. All regression assumptions
(normality, constant variance, independence of the error term) hold.
slide 8
Drivers of Oil Price other than US Oil Consumption
DSCI 5180
Decision Making
The same drivers may partially affect Inflation-Adjusted Oil Price. We fit a
second regression model that explains US Oil Price (adjusted for
inflation). The unexplained part (residuals) is a US Oil Price Differential.
Inflation-Adjusted US Oil Price = f(World Oil Production, US Population,
US Consumption)
+ US Oil Price Differential
slide 9
Price vs. Consumption after the model adjustments
DSCI 5180
Decision Making
 Plotting Residuals1 (=US Oil Price Differential) vs.
Residuals2 (=US Oil Consumption Differential) reveals a
shape that is much closer to a demand curve
slide 10
Adding a quadratic demand curve
DSCI 5180
Decision Making
Transfer your data
to Excel, plot the
scatterplot, and
then add a
trendline. Change
the trendline
settings to a
second-order
polynomial curve
slide 11
Final measurement scale adjustment
DSCI 5180
Decision Making
The plot shown in the previous slide uses “differential”
measurement scales. These may be hard to interpret.
Adding a constant to all data for the two variables
would not alter their relationship or the shape of the
curve. Add the average price to all US Oil Price
Differential values and add the average consumption
value to all US Oil Consumption Differential values
Model-Adjusted US Oil Price = US Oil Price Differential
+ Average (Price)
Model-Adjusted US Oil Consumption = US Oil Consumption Differential
+ Average (Consumption)
slide 12