Prediction with limited information in automatic Valuation Systems

Dimitrios Papastamos, Dimitris Karlis, Angelos Kekempanos and Antonios Alexandridis

ERES from European Real Estate Society (ERES)

Abstract: Automatic valuation systems have found several applications in real estate market. It is common to build such systems on historical data aiming at predicting new properties in a comprehensive manner. In recent days it is common that the data are collected not based on valuations but from other sources and, hence, they can be incomplete. The paper aims at examining the effect of such incomplete data in the prediction accuracy. We apply a nearest neighbor imputation technique to see how the proportion of missing information can affect the results but also to reveal for separate variables their impact and importance in the derivations. We make use of a real valuation data set from a Greek financial institution, and we see how much the proportion of missingness impacts the results but also that there are variables that when missing it is very difficult to impute them.

Keywords: error; imputation; Prediction (search for similar items in EconPapers)
JEL-codes: R3 (search for similar items in EconPapers)
Date: 2022-01-01
New Economics Papers: this item is included in nep-ure
References: Add references at CitEc
Citations:

Downloads: (external link)
https://eres.architexturez.net/doc/eres-id-eres2022-211 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:arz:wpaper:2022_211

Access Statistics for this paper

More papers in ERES from European Real Estate Society (ERES) Contact information at EDIRC.
Bibliographic data for series maintained by Architexturez Imprints ().