This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: ECON 2P91: Business Econometrics with Applications Fall 2009 Assignment 1 SOLUTIONS Consider the Excel data file princegeorge_housing.xls . In this assignment, you will investigate the relationship between housing price (PRICE) and lot size (LOT). Do houses with bigger lot sizes sell for higher prices? To answer this question, we will construct a simple hedonic price model relating the price of a residential house to its characteristics (a hedonic price model relates the price of an underlying asset (a residential house in this case) to its characteristics). Although other characteristics (e.g., number of bedrooms, number of bathrooms, etc.) might be important, only one characteristic, LOT, will be considered in this assignment. You will run a regression of PRICE on LOT using data for the 1011 residential houses that were sold in the City of Prince George, British Columbia, Canada, during the period from the first quarter of 2001 to the first quarter of 2002. For this data set, LOT is measured in acres and PRICE is measured in thousands of Canadian dollars; QT is the quarter in which each house was sold (QT=1 denotes the first quarter (winter); QT=2 denotes the second quarter (spring); QT=3 denotes the third quarter (summer); and QT=4 denotes the fourth quarter (fall)). The variable ID is the housing identification variable. 1. Create a scatterplot using LOT and PRICE. Put PRICE on the vertical axis and LOT on the horizontal axis. Note that GRETL will also overlay the regression line on the scatterplot. Comment on the possible relationship between lot size and residential housing prices. 0 50 100 150 200 250 300 350 400 0 10 20 30 40 50 60 70 80 90 price lot price versus lot (with least squares fit) Y = 118. + 1.35X Comment: Scatter diagram shows positive relationship between LOT and PRICE. 2. Compute the correlation coefficient between LOT and PRICE. Use the estimated correlation coefficient to comment on the possible relationship between lot size and residential housing prices. What are the units of the correlation coefficient? corr(price, lot) = 0.20227717 Under the null hypothesis of no correlation: t(1009) = 6.56091, with two-tailed p-value 0.0000 Correlation coefficient = 0.2. The positive correlation indicates positive relationship between LOT and PRICE. However, the correlation coefficient is close to zero, indicating a weak positive relationship. The correlation coefficient is units free, indicating that the value is independent of the units in which the variables are measured i.e. We could still express PRICE in terms of dollars (instead of thousands of dollars) and end up with the same correlation coefficient value....
View Full Document
- Fall '09