MONASH UNIVERSITY
ECONOMETRICS AND BUSINESS STATISTICS
Applied Econometrics
Tutorial 2
1. Let yi, xi, wi be three i.i.d. random variables with mean zero and variance equal to 𝜎 ,
𝜎 and 𝜎 , respectively. Suppose also that 𝐸(𝑦 𝑥 ) = µ , 𝐸(𝑦 𝑤 ) = µ and
𝐸(𝑤 𝑥 ) = µ .
Compute 𝑐𝑜𝑣[(𝑦 − 𝑥 ), (𝑤 − 𝑥 )].
2. Page 299, Ex. 6.8. parts a, b, c, d, e. (Carter, Griffiths, Lim - Principles of Econometrics,
5th Ed. (2017))
1
3. Practice on EViews and learn about some of the interpretation of indicator variables:
Use the utown.wfl dataset to study the effect of living near a major University (called
University Town) on house prices. The three variables in the data set that you will need
are price (measured in $1000 units, utown (an indicator that takes the value of 1 if the
house is located near the university) and SQFT (house size, measured in square feet).
a. Estimate the model 𝑃𝑅𝐼𝐶𝐸 = 𝛼 +𝛼 𝑈𝑇𝑂𝑊𝑁 + 𝑒 and hence estimate the house
price premium for living near the university.
b. Create an indicator variable DTOWN (downtown) for houses that are not near the
university by defining DTOWN = 1 - UTOWN, and then estimate the model
𝑃𝑅𝐼𝐶𝐸 = 𝛽 +𝛽 𝐷𝑇𝑂𝑊𝑁 + 𝑒. Use this model to estimate the discount for housing
that is not near the university.
c. Estimate the model 𝑃𝑅𝐼𝐶𝐸 = 𝛾 𝐷𝑇𝑂𝑊𝑁 +𝛾 𝑈𝑇𝑂𝑊𝑁 + 𝑒 (this model has no
intercept). Use this model to estimate the mean prices for downtown houses and
uptown houses.
d. Comment on why the 𝑅 and SSE measures of fit for each of these models in a., b.
and c. are the same.
e. Estimate the model 𝑃𝑅𝐼𝐶𝐸 = 𝛿 + 𝛿 𝐷𝑇𝑂𝑊𝑁 + 𝛿 𝑈𝑇𝑂𝑊𝑁 + 𝑒 and then
comment on what happens.
f. Estimate the model 𝑃𝑅𝐼𝐶𝐸 = 𝜇 + 𝑒. Interpret your estimated parameter 𝜇̂ .
Then determine the proportions of UTOWN and DTOWN houses in the sample,
and hence find a relationship between 𝜇̂ , 𝛾 and 𝛾 .
g. Estimate and compare the two estimated equations
𝑃𝑅𝐼𝐶𝐸 = 𝛼 +𝛼 𝑈𝑇𝑂𝑊𝑁 + 𝛼 𝑆𝑄𝐹𝑇 + 𝑒
𝑃𝑅𝐼𝐶𝐸 = 𝛽 +𝛽 𝐷𝑇𝑂𝑊𝑁 + 𝛽 𝑆𝑄𝐹𝑇 + 𝑒.
