Efficacy of Hedonic Estimation with the Annual Housing Survey: Evidence From the Demand Experiment. (RE-ANNOUNCEMENT of PB85-127074-see notes field for explanation)

摘要

The overall purpose of this study is to determine whether the lack of certain location and structural information in the Annual Housing Survey reduces its usefulness as a data source for hedonic estimation (mathematical device by which distinctive features of housing units, such as room size and number, are reduced to quantities that reflect their relative importance in the housing market). Specifically, the study inquires into whether variables collected by the Housing Allowance Demand Experiment, but not the Annual Housing Survey, make important improvements in explaining rent and how omission of variables from the Annual Housing Survey alters the estimated contribution to rent variables included in the survey. The paper first describes the methodology used to address the problem of omitted variables, and then the two data sets are explained in greater detail as well as the justiciation for using Demand Experiment data to make inferences about the Annual Housing Survey. In determining whether variables collected by the Demand Experiment, but not the Annual Housing Survey, make important improvements in explaining rent, the study concludes that the variables are very likely to make a statistically significant contribution to rent in the Annual Housing Survey. However, the addition of the entire set of omitted variables to the Annual Housing Survey does not appear to make an important addition to the explanatory power of an hedonic model. Other issues examined include (1) how omission of variables from the Annual Housing Survey alters the estimated contribution to rent of variables included in the survey, (2) bias in the size of coefficients, (3) bias in the variance of coefficients, (4) increased stability in the complete model, and (5) evaluating the trade - off between bias and stability. Tabular data and 18 notes are provided.