An assessment of property performance forecasts: consensus versus econometric

Published date01 September 2006
Pages386-399
Date01 September 2006
DOIhttps://doi.org/10.1108/14635780610691896
AuthorSotiris Tsolacos
Subject MatterProperty management & built environment
An assessment of property
performance forecasts: consensus
versus econometric
Sotiris Tsolacos
Property and Portfolio Research, London, UK
Abstract
Purpose The paper seeks to evaluate accuracy and efficiency of consensus forecasts for all
property rents and total returns in the UK. The aim of the paper is to investigate whether consensus
forecasts, containing a high degree of judgement, are better than forecasts produced by uncomplicated
time-series and econometric models that practitioners can easily estimate and use for forecasting.
Design/methodology/approach – This study estimates simple models of all property rents and
returns and generates forecasts for one- and two-year horizons on a rolling basis over the period 1999
to 2004. These forecasts are real time forecasts. That is they are made using information available to
the analyst at the time of the forecast each year and no future knowledge is assumed. The forecasts
made by these models are compared with the corresponding consensus forecasts. The comparative
assessment is based on conventional tests for bias, variability and efficiency of forecasts.
Findings – If attention is focused on rents, the consensus forecast is ranked best for the one-year
horizon but it is outperformed by the regression model and a combination of the statistical models for
the two-year horizon. For the one-year and two-year forecasts of total returns a simple regression
model with interest rates clearly improves upon the consensus forecasts. There is clear evidence that
the consensus forecasts fail to incorporate the information contained in recent interest rate movements.
Therefore subject to the sample period for this analysis the survey forecasts of total returns cannot be
considered impartial. Analysts should include base rate information into their predictions.
Originality/value – This is one of the few attempts to formally evaluate consensus forecasts in the
real estate field and perform a direct comparison with quantitative forecasts. It produces initial
evidence suggesting that highly judgemental consensus forecasts do not necessarily outperform
quantitative forecasts based on fundamentals. It prompts property forecasters and investors to engage
in forecast evaluation and include missing information and past errors in their forecasts.
Keywords Rent reviews, Returns,Econometrics, Comparativetests, Forecasting
Paper type Research paper
1. Introduction
Property forecasts are now widely used and increasingly receive more weight in real
estate investment analysis and portfolio strategy. The perceived need of property
investors for formal forecasting underpins the greater engagement of the property
profession in this field. Property investors assess both quantitative and qualitative
evidence to arrive at their final decisions. Quantitative forecasting plays an important
role since future outcomes are conditional on identifiable drivers and their future
trends. Moreover, investors see formal forecasting approaches complement ing
qualitative forecasting. This interest has resulted in the construction of forecast
models, ranging from simple techniques to more advanced methodologies with
stationary data, and has led to a greater availability of forecasts.
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1463-578X.htm
JPIF
24,5
386
Received February 2006
Accepted June 2006
Journal of Property Investment &
Finance
Vol. 24 No. 5, 2006
pp. 386-399
qEmerald Group Publishing Limited
1463-578X
DOI 10.1108/14635780610691896

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT