Assignment 2 (2015F)

CONCORDIA UNIVERSITY
Department of Economics
ECON 222/2 SECTIONS A and B
STATISTICAL METHODS II
FALL 2015 ASSIGNMENT 2
Due: Monday, November 9, before 3:00 pm
1. (14 marks) We have defined the simple linear regression model to be y = 1 + 2 x + e .
Suppose however that we knew, for a fact, that 1 = 0 .
a. (2 marks) What does the linear regression model look like, algebraically, if 1 = 0 ?
b. (2 marks) What does the linear regression model look like, graphically, if 1 = 0 ?
c. (2 marks) If 1 = 0 , the least squares sum of squares function becomes

N
S ( 2 ) = ( yi 2 xi ) . Using the data

i=1
x 1 2 3 4 5 6
y 4 6 7 7 9 11
Plot the value of the sum of squares function for enough values of 2 for you to locate the
approximate minimum. What is the significance of the value of 2 that minimises S ( 2 ) ?
d. (2 marks) Using calculus, show that the formula for the least squares estimate of 2 in
N
N
this model is b2 = xi yi xi2 . Use this result to compute b2 and compare this
i=1
i=1
value to the value you obtained geometrically.
e. (2 marks) Using the estimate obtained with the formula in part (d), plot the fitted
(estimated) regression function. On the graph locate the point ( x, y ) . What do you
observe?
f. (2 marks) Using the estimates obtained with the formula in part (d), obtain the least
squares residuals, ei = yi b2 xi . Find their sum.
g. (2 marks) Calculate
x e .
i i
i=1
2. (12 marks) Using the regression output for the food expenditure model shown in Figure 2.9,
a. (2 marks) Construct a 95 percent interval estimate for 1 and interpret.
b. (2 marks) Test the null hypothesis that 1 is zero against the alternative that it is not at
the 5 percent level of significance without using the reported p-value. What is your
conclusion?
c. (2 marks) Draw a sketch showing the p-value 0.0622 shown in Figure 2.9, the critical
value from the t-distribution used in part (b), and how the p-value could have been used to
answer part (b).
d. (2 marks) Test the null hypothesis that 1 is zero against the alternative that it is positive
at the 5 percent level of significance. Draw a sketch of the rejection region and comput
the p-value. What is your conclusion?

e. (2 marks) Explain the differences and similarities between the level of significance and
the level of confidence.
f. (2 marks) The results in part (d) show that we are 95 percent confident that 1 is positive.
True or false? If false, explain.
3. (4 marks) Suppose you are estimating a simple linear regression model.

a. (2 marks) If you multiple all the x values by 20, but not the y values, what happens to the
parameter values 1 and 2 ? What happens to the least squares estimates b1 and b2 ?
What happens to the variance of the error term?
b. (2 marks) Instead, if you multiple all the y values by 50, but not the x values, what
happens to the parameter values 1 and 2 ? What happens to the least squares estimates
b1 and b2 ? What happens to the variance of the error term?
4. (14 marks) The file stockton2.dat contains data on 880 houses sold in Stockton, California,
during mid-2005. Variable descriptions are in the file stockton2.def.
a. (2 marks) Estimate the log-linear model ln PRICE = 1 + 2 SQFT + e . Interpret the
estimate model parameters. Calculate the slope and elasticity at the sample means, if
necessary.
b. (2 marks) Estimate the log-log model ln PRICE = 1 + 2 ln SQFT + e . Interpret the

estimate model parameters. Calculate the slope and elasticity at the sample means, if
necessary.
c. (2 marks) Compare the R 2 -value from the linear model PRICE = 1 + 2 SQFT + e to the
generalised R 2 measure for the models in parts (b) and (c).
d. (2 marks) Construct histograms of the least squares residuals from each of the models in
parts (a), (b) and (c) and obtain the Jarque-Bera statistics. Based on your observations, do
you consider the distributions of the residuals to be compatible with an assumption of
normality?
e. (2 marks) For each of the models in parts (a) (c), plot the least squares residuals against
SQFT. Do you observe any patterns?
f. (2 marks) For each model in parts (a) (c), construct a 95 percent prediction interval for
the value of a house with 2700 square feet.
g. (2 marks) Based on your work in this problem, discuss the choice of function form.
Which functional form would you use? Explain.

Assignment 2 (2015F)

Uploaded by

Copyright:

Available Formats

Assignment 2 (2015F)

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Assignment 2 (2015F)

Uploaded by

Copyright:

Available Formats

CONCORDIA UNIVERSITY

c. (2 marks) If 1 = 0 , the least squares sum of squares function becomes

S ( 2 ) = ( yi 2 xi ) . Using the data

3. (4 marks) Suppose you are estimating a simple linear regression model.

b. (2 marks) Estimate the log-log model ln PRICE = 1 + 2 ln SQFT + e . Interpret the

You might also like