(1) PRCE, =+,5QFT1+ (a) E(c)=, (3) E(e2)=2 and (4) E(,i)=0 fot t=z where t PRICE denotes the sale price and t SQFT denotes the floor area of house t. The EViews workfile br.wf1 contains data on the prices and floor areas of 1080 houses sold in Baton Rouge, Louisiana, in mid2005. a) Explain what is meant by each of the assumptions (2) to (4). b) Use ordinary least squares to estimate the unknown parameters. Write down the estimated model (including standard errors). c) Interpret the estimates b1 and b2. Are the signs of these estimates consistent with your expectations? d) Use a t-test to test H0:2=80 at the 5% level of significance. e) What does the Gauss-Markov theorem tell us about the ordinary least squares estimators? f) For this part, assume the observations in the data file constitute a population, and consider four samples of size 10 from this population: Sample 1 = observations 110 Sample 2 = observations 4150 Sample 3 = observations 501.510 Sample 4= observations 721730 Use each of these four samples to estimate the model. What do you notice about the values of b1 and b2 obtained using each of the four samples? What do you notice about the average values of b1 and b2 over the four samples? g) For this part, again assume the observations in the data file constitute a population, but this time consider three samples of size 150 : Sample 1 = observations 151300 Sample 2 = observations 451600 Sample 3 = observations 801950 Use each of these three samples to estimate the model. What do you notice about the estimates obtained using these three samples compared with the estimates obtained in part f)?.

1. (1) PRCE, =+,5QFT1+ (a) E(c)=, (3) E(e2)=2 and (4) E(,i)=0 fot t=z where t PRICE denotes
the sale price and t SQFT denotes the floor area of house t. The EViews workfile br.wf1 contains
data on the prices and floor areas of 1080 houses sold in Baton Rouge, Louisiana, in mid2005. a)
Explain what is meant by each of the assumptions (2) to (4). b) Use ordinary least squares to
estimate the unknown parameters. Write down the estimated model (including standard errors).
c) Interpret the estimates b1 and b2. Are the signs of these estimates consistent with your
expectations? d) Use a t-test to test H0:2=80 at the 5% level of significance. e) What does the
Gauss-Markov theorem tell us about the ordinary least squares estimators? f) For this part,
assume the observations in the data file constitute a population, and consider four samples of size
10 from this population: Sample 1 = observations 110 Sample 2 = observations 4150 Sample 3 =
observations 501.510 Sample 4= observations 721730 Use each of these four samples to estimate
the model. What do you notice about the values of b1 and b2 obtained using each of the four
samples? What do you notice about the average values of b1 and b2 over the four samples? g)
For this part, again assume the observations in the data file constitute a population, but this time
consider three samples of size 150 : Sample 1 = observations 151300 Sample 2 = observations
451600 Sample 3 = observations 801950 Use each of these three samples to estimate the model.
What do you notice about the estimates obtained using these three samples compared with the
estimates obtained in part f)?