ANALYZING REAL-ESTATE DATA PROBLEMS USING THE GIBBS SAMPLER

Real estate data are often characterized by data irregularities: missi ng data, censoring or truncation, measurement error, etc. Practitioner s often discard missing- or censored-data cases and ignore measurement error. We argue here that an attractive remedy for these irregularity problems is simulation-based model fitting using the Gibbs sampler. T he style of the paper is primarily pedagogic, employing a simple illus tration to convey the essential ideas, unobscured by implementation co mplications. Focusing on the missing-data problem, we show dramatic im provement in inference by retaining rather than deleting cases of part ially observed data. We also detail Gibbs-sampler usage for other data problems.