Robust estimation of hedonic models of price and income for investment property
| DOI | https://doi.org/10.1108/EUM0000000005789 |
| Date | 01 August 2001 |
| Pages | 342-360 |
| Published date | 01 August 2001 |
| Author | Christian Janssen,Bo Söderberg,Julie Zhou |
| Subject Matter | Property management & built environment |
JPIF
19,4
342
Journal of Property Investment &
Finance, Vol. 19 No. 4, 2001,
pp. 342-360. #MCB University
Press, 1463-578X
ACADEMIC PAPERS
Robust estimation of hedonic
models of price and income for
investment property
Christian Janssen
Faculty of Business, University of Victoria, Victoria, Canada
Bo SoÈderberg
Real Estate Economics, Royal Institute of Technology,
Stockholm, Sweden, and
Julie Zhou
Department of Mathematics and Statistics, University of Victoria,
Victoria, Canada
Keywords Income, Capitalization, Estimating
Abstract Real estate market data often contain outliers in the observations. Since outliers have
a large influence on least squares estimates, robust regression methods have been recommended
for this situation. Compares the performance of least squares and least median of squares, a
robust method, in the estimation of price/income relationships for apartment buildings.
Multiplicative models with multiplicative errors are estimated by means of natural log
transformations. The study confirms the importance of employing robust methods for this
application and implies this may well be so for real estate data sets more generally.
1. Introduction
Valuation of real estate is a worldwide phenomenon. A wide range of properties
are valued for a multitude of purposes. For residential property, the more
common are purchase and sale, transfer, tax assessment, expropriation,
divorce, inheritance or estate settlement, investment and financing. For
commercial property there are also mergers and acquisitions and the
preparation of annual reports. The valuations are generally performed by
members of different professions ± real estate agents, appraisers, assessors,
mortgage lenders, or other specialists and consultants. Occasionally
adjudication or arbitration by boards, tribunals, courts and other judicial and
administrative bodies is involved. The institutional set up, regulations, and
practices vary between jurisdictions, countries and regions of the world, but the
predominant focus in the valuations is the estimation of market value.
The market prices of income properties depend primarily on the income
generated, with price being some multiple of income. A number of different
The research register for this journal is available at
http://www.mcbup.com/research_registers
The current issue and full text archive of this journal is available at
http://www.emerald-library.com/ft
Helpful comments by two anonymous reviewers are gratefully acknowledged. The research of
Julie Zhou is supported by the Natural Sciences and Engineering Research Council of Canada.
Academic
papers:
Hedonic models
343
valuations are performed this way. Stocks are valued based upon price/
earnings ratios, bonds based upon effective yield to maturity, businesses based
upon earnings, multipliers and cash flows generally based upon discounting.
The predominant approach to the valuation of income property is capitalization
of the income stream, that is, division by a capitalization rate (Parker, 1996).
What the cap rate ought to be in specific cases is a difficult question. It is
affected by a number of factors such as the composition of the income, its long
run stability, expectations of future growth, opportunities for capital
appreciation, operating and maintenance costs, as well as risk considerations.
Economic conditions also exert a major influence. Investors generally differ in
their assessment of the net effect of influencing factors, as well as in their
attitude toward risk. This leads to properties being sold by some and
purchased by others. As properties are traded every day it is evident that
decisions are made with respect to appropriate capitalization rates. It is not
possible to deduce what may have been driving forces in individual
transactions, but inferences can be made from the resulting market prices with
respect to the effects of major property characteristics in the market as a whole.
Price depends on income, but income depends on the features of the
property. This creates a relationship between market prices on the one hand
and property characteristics on the other. The relationships between income
and characteristics, and between price and characteristics permit inferences
with respect to that between the capitalization rate and characteristics.
We investigate these relationships in this paper. We first study that between
income and features, and then that between price and features. Inferences are
then made for the cap rate, the ratio of income to price. We are interested in
gaining insight into how key variables are considered by the market of
landlords and tenants (the rental market) and the market of buyers and sellers
(the property market) in the valuation of income property. We are especially
interested in features considered differently in the two markets, as well as in
the factors that appear to affect the cap rate used.
A major difficulty with real estate data is that there are often extraneous
circumstances surrounding the sale that are not recorded in the data system.
For residential real estate, for instance, divorce is a cause of outliers, as interest
in holding out for a higher price may yield to the pressure of getting the
property sold. A property sold by an estate or by siblings for the purpose of
dividing the proceeds may obtain a below-market price. For commercial real
estate, a legitimate arm's-length transaction may be motivated by the vendor
needing to liquidate some holdings as a result of other business engagements.
For instance, not all of Volvo Passenger Car's 2.4 million square meters of space
in Europe might be needed after its acquisition by Ford. In the disposition of
excess properties, achieving efficiency in business operations would be a
higher priority than obtaining a certain price for individual properties.
Extraneous pressure to sell, or motivation to sell, may result in a lower price
than normal. Some countries permit immigration in an ``investor'' category by
To continue reading
Request your trial