Details
| Translated title of the contribution | Possibilities and limits of integration capability of heterogenous data for the valuation in real areas with few transactions |
|---|---|
| Original language | German |
| Pages (from-to) | 247-258 |
| Number of pages | 12 |
| Journal | AVN Allgemeine Vermessungs-Nachrichten |
| Volume | 126 |
| Issue number | 10 |
| Publication status | Published - 2019 |
Abstract
Comprehensive market data for all functional and spatial submarkets are the basis for a reliable and accurate property valuation. In general, these data are derived from a sufficient number of purchase prices using statistical approaches (hedonic models) and published in market reports. These approaches apply whenever there is a sufficient number of purchase cases for a spatial and functional submarket. If the number of transactions is low (so-called regions with few transactions), classical statistical approaches fail to derive the market data and, as a result, it thus became more difficult to determine the market value. As a result of the high lack of transparency in these markets, there are many unusual cases of purchase (outliers), i. e. cases that deviate strongly from the average market. However, to derive market data (e. g. in the form of comparison factors) in regions with few transactions, there is a need for research into the development of methods for these special markets: In particular, the ability to integrate further information (offer data, compiled appraisals, expert knowledge) into the evaluation is in the foreground here. This paper uses the application of a robust Bayesian model to derive market data from a combination of expert knowledge and purchase price data. In /Dorndorf et al. 2016/ an approach for the evaluation of regions with few transactions in a simulated environment has already been developed. In this article, this methodology is applied and validated for the first time in a real area with few transactions. In addition, findings on data sources and data quality are presented and discussed. The results of the study show that the methodology developed can be used to integrate additional market data into real estate valuations. However, this requires a significantly increased effort in collecting the additional data. Nevertheless, the methodology presented provides an opportunity for valuation of regions with few transactions.
ASJC Scopus subject areas
- Social Sciences(all)
- Geography, Planning and Development
- Engineering(all)
- Civil and Structural Engineering
- Earth and Planetary Sciences(all)
- Earth and Planetary Sciences (miscellaneous)
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: AVN Allgemeine Vermessungs-Nachrichten, Vol. 126, No. 10, 2019, p. 247-258.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Möglichkeiten und Grenzen der Integrationsfähigkeit unterschiedlicher Daten für die Bewertung in realen kaufpreisarmen Lagen
AU - Soot, Matthias
AU - Dorndorf, Alexander
AU - Alkhatib, Hamza
AU - Weitkamp, Alexandra
N1 - Publisher Copyright: © 2019, VDE VERLAG GMBH. All rights reserved. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.
PY - 2019
Y1 - 2019
N2 - Comprehensive market data for all functional and spatial submarkets are the basis for a reliable and accurate property valuation. In general, these data are derived from a sufficient number of purchase prices using statistical approaches (hedonic models) and published in market reports. These approaches apply whenever there is a sufficient number of purchase cases for a spatial and functional submarket. If the number of transactions is low (so-called regions with few transactions), classical statistical approaches fail to derive the market data and, as a result, it thus became more difficult to determine the market value. As a result of the high lack of transparency in these markets, there are many unusual cases of purchase (outliers), i. e. cases that deviate strongly from the average market. However, to derive market data (e. g. in the form of comparison factors) in regions with few transactions, there is a need for research into the development of methods for these special markets: In particular, the ability to integrate further information (offer data, compiled appraisals, expert knowledge) into the evaluation is in the foreground here. This paper uses the application of a robust Bayesian model to derive market data from a combination of expert knowledge and purchase price data. In /Dorndorf et al. 2016/ an approach for the evaluation of regions with few transactions in a simulated environment has already been developed. In this article, this methodology is applied and validated for the first time in a real area with few transactions. In addition, findings on data sources and data quality are presented and discussed. The results of the study show that the methodology developed can be used to integrate additional market data into real estate valuations. However, this requires a significantly increased effort in collecting the additional data. Nevertheless, the methodology presented provides an opportunity for valuation of regions with few transactions.
AB - Comprehensive market data for all functional and spatial submarkets are the basis for a reliable and accurate property valuation. In general, these data are derived from a sufficient number of purchase prices using statistical approaches (hedonic models) and published in market reports. These approaches apply whenever there is a sufficient number of purchase cases for a spatial and functional submarket. If the number of transactions is low (so-called regions with few transactions), classical statistical approaches fail to derive the market data and, as a result, it thus became more difficult to determine the market value. As a result of the high lack of transparency in these markets, there are many unusual cases of purchase (outliers), i. e. cases that deviate strongly from the average market. However, to derive market data (e. g. in the form of comparison factors) in regions with few transactions, there is a need for research into the development of methods for these special markets: In particular, the ability to integrate further information (offer data, compiled appraisals, expert knowledge) into the evaluation is in the foreground here. This paper uses the application of a robust Bayesian model to derive market data from a combination of expert knowledge and purchase price data. In /Dorndorf et al. 2016/ an approach for the evaluation of regions with few transactions in a simulated environment has already been developed. In this article, this methodology is applied and validated for the first time in a real area with few transactions. In addition, findings on data sources and data quality are presented and discussed. The results of the study show that the methodology developed can be used to integrate additional market data into real estate valuations. However, this requires a significantly increased effort in collecting the additional data. Nevertheless, the methodology presented provides an opportunity for valuation of regions with few transactions.
KW - Areas with few transactions
KW - Real estate valuation
KW - Robust Bayesian regression
UR - http://www.scopus.com/inward/record.url?scp=85074127683&partnerID=8YFLogxK
M3 - Artikel
AN - SCOPUS:85074127683
VL - 126
SP - 247
EP - 258
JO - AVN Allgemeine Vermessungs-Nachrichten
JF - AVN Allgemeine Vermessungs-Nachrichten
SN - 0002-5968
IS - 10
ER -