The role of apartment living area and floor level in hedonic price formation across market segments

Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

School of Business | Master's thesis

Authors

Date

2024

Department

Rahoituksen laitos

Major/Subject

Mcode

Degree programme

Finance

Language

en

Pages

75 + 5

Series

Abstract

The purpose of this thesis is the examination of price segmentation in the Finnish housing markets, and the price formation within the segments using hedonic price modeling, with further emphasis placed on living area and apartment floor level. The examination includes the overall price impact of the characteristics of interest in the full market sample and in price-segmented subsamples. Subsample OLS regressions combined with Chow and Tiao-Goldberger F-tests are used to determine the potential for price segmentation. In addition, quantile regression estimates are used as a robustness check for the results provided by subsample regressions. Sample of apartment transactions from the largest cities in Finland from 2003 until 2020 was utilized in the model estimation, provided by the Central Federation of Finnish Real Estate Agencies (Kiinteistönvälitysalan Keskusliitto, KVKL). The subsample results indicate that Finnish housing markets are price-segmented. Chow and Tiao-Goldberger tests found highly statistically significant differences between the regression coefficients of low- and high-priced segments. Furthermore, the effect of living area on the whole market sample was as expected, and participants in the high-priced segment seem to value living area higher, with strong evidence from both OLS and quantile regression estimates. However, inconsistent results were found for the apartment floor. Flexible functional specification, namely floor level dummy variable form, revealed statistically significant interaction with elevator access in the whole market sample. In addition, non-linear price dynamics were found. Examination of the price segment dynamics using subsample OLS regressions provided supporting evidence for increased willingness to pay for higher floors in the high-valued segment. However, quantile regression estimates contradicted these. In fact, the difference between segments seems to be inverse from the hypothesized, although consistent significant differences were not found, except for individual floors.

Description

Thesis advisor

Suominen, Matti

Keywords

residential real estate, hedonic regression, quantile regression, price segmentation

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202403172694

Collections

[gradu] Kauppakorkeakoulu / BIZ