Housing Price Data
Housing Price Data, used by Karl Case and Robert Shiller in their papers: "Prices of Single-Family Homes since 1970: New Indexes for Four Cities," New England Economic Review, Sept/Oct 1987 pp. 45-56, and "The Efficiency of the Market for Single Family Homes," American Economic Review 1989.
There are four data files, one for each city.
Raw data on individual houses are double sales of individual houses. In each row of the matrix, the first observation is the first sale price, the second observation is the second sale price, the third is the quarter of the first sale (1970-1 = 1) and the fourth is the quarter of the second sale. For Atlanta and Dallas, the final zero on the price is omitted, for Chicago and Oakland, the final two zeros on the price are omitted. The order of the observations is not random. The first part of the file consists of observations where the price changes are positive. Then, there follow negative price changes, and finally, after the word @outliers:@, those observations that were judged to be outliers and which were deleted before our data analysis.
The outliers were deleted because the price change appeared to be in error, on no more information than the price change itself. Often, the price change was in the vicinity of a factor of ten, suggesting a misplaced decimal in the data.