Class Project
You will estimate a hedonic model Chicago and New Orleans by using AHS 2009 metro data. you should pick Chicag. The SMSA code for Chicago is 7600. Since you will work on a single year data, you will estimate the following equation;
Ln (Pi) = β0+ Hiβ1 + Siβ2 + Niβ3 + εi,
This equation is similar to equation (1) in Downes and Zabel (2002)(in attachment). Year subscripts drop because we focus on only one year, namely 2009 in AHS metro data. Table 1 presents the list of variables that you will use in your estimation. First you will create all the variables listed under table 1. Then you will create a summary table for these variables where you will report the mean and standard deviation for each variable. Please read Downes and Zabel (2002) paper very carefully. They have census tract level neighborhood data and you will create your neighborhood level data at ZONE level for Chicago. For New Orleans you can use census tract or zone level data depending on availability.
Table 1
Variable names, definitions, and summary statistics
Variable Definition Mean Std. dev.
LNVALUE Natural log of owner-estimated house value:
CENCITY 1 if house in central city of SMSA, 0 otherwise
HAGE The age of the house in years
HAGESQ/100 The square of the age of the house in years/100:
GARAGE 1 if the house has a garage, 0 otherwise
BEDROOMS Number of bedrooms in the house
FULLBATHS Number of full bathrooms in the house:
AIRCOND 1 if the house has either central or room air
conditioning, 0 otherwise
TENURE 1 if the household moved into the unit in the12 months prior to survey, 0 otherwise
TENURE510 1 if household in unit 5 to 10 years, 0 otherwise
LNMEDI Natural log of median income in the house's zone in tens of thousands of dollars
MEDAGE Median age of individuals in the house's zone
PROPHS Proportion of individuals over 25 who have completed
high school in the zone
PROPVAC Proportion of housing units vacant in the zone
PROPLO Proportion of houses in the zone with less than
one occupant per room
AFRAM Proportion African American population in the zone
HISPNA Proportion of hispanic or native American population in the zone
PROPRENT Proportion of rental units in the zone
PROPSCAGE Proportion of population that is school aged in the zone
DSchool=1 if residents are satisfied with the public schools in the zone, 0 otherwise
Once you create your summary table, then you estimate your model by using these variables.
Table 2 will present mean and standard deviation of all your variables. Table 3 should report the correlations among the variables (similar to Table 2 in Downes and Zabel (2002).
Table 4 should present your results. Make sure that you will report regression coefficients, standard errors, R2, and number of observations, significance levels (p-stat) and Ftest for the whole model. Note you need to use weight in all your estimations and Tables 2, 3, and 4. Make sure that you cluster you standard errors at zone level.
In analysis and discussion section you should discuss how housing characteristics, neighborhood characteristics and school quality are capitalized in housing prices.