Dwelling price ranking versus socioeconomic clustering: Possibility of imputation

Larisa Fleishman, Yury Gubman, Aviad Tur-Sinai

Research output: Contribution to journalArticlepeer-review

Abstract

In order to characterize the socioeconomic profile of various geographic units, it is common practice to use aggregated indices. However, the process of calculating such indices requires a wide variety of variables from various data sources available concurrently. Using a number of administrative databases for 2001 and 2003, this study examines the question of whether dwelling prices in a given locality can serve as a proxy for its socioeconomic level. Based on statistical and geographic criteria, we developed a Dwelling Price Ranking (DPR) methodology. Our findings show that the DPR can serve as a good approximation for the socioeconomic cluster (SEC) calculated by the Israel Central Bureau of Statistics for years when the required data was available. As opposed to the SEC, the suggested DPR indicator can easily be calculated, thus ensuring a continuum of socioeconomic index series. Both parametric and nonparametric statistical analyses have been carried out in order to examine the additional social, demographic, location, crime and security effects that are exogenous to SEC. Complementary analysis on recently published SEC series for 2006 and 2008 show that our conclusions remain valid. The proposed methodology and the obtained findings may be applicable for different statistical purposes in other countries which possess dwelling transactions data.

Original languageEnglish
Pages (from-to)205-229
Number of pages25
JournalJournal of Official Statistics
Volume31
Issue number2
DOIs
StatePublished - 1 Jun 2015
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2015 Statistics Sweden.

Keywords

  • Housing market
  • Index construction
  • Urban locality

ASJC Scopus subject areas

  • Statistics and Probability

Fingerprint

Dive into the research topics of 'Dwelling price ranking versus socioeconomic clustering: Possibility of imputation'. Together they form a unique fingerprint.

Cite this