Use this URL to cite or link to this record in EThOS:
Title: Modelling and predicting commercial property performance indicators : theoretical and practical aspects
Author: Chaplin, R. A.
Awarding Body: University of Cambridge
Current Institution: University of Cambridge
Date of Award: 1998
Availability of Full Text:
Full text unavailable from EThOS.
Please contact the current institution’s library for further details.
This thesis examines various aspects of the modelling and prediction of Commercial Property Performance Indicators (CPPIs) with special reference to the modelling of office rents carried out by academics and practitioners. It considers the appropriate strategy for investment in commercial property, given that the efficiency of the market at reflecting available information into prices is largely unknown. It is argued and indeed accepted wisdom that it is, to some extent, inefficient and some investors appear consistently to 'beat the market'. Out of the uncertainty as to the efficiency of the market, CPPIs are modelled, predicted and forecast by academics and practitioners in an attempt to explain future price movements and thus provide knowledge which can be exploited to earn abnormal gains. Modelling and predicting commercial property markets (and so investment) is complicated by the nature of available CPPIs, being the results of valuations of properties rather than transactions. This means that CPPIs may not accurately reflect movements in the market but rather they may lag and underestimate actual changes - a phenomenon known as 'smoothing'. A model is presented which can be used to unsmooth CPPIs using a multiple regime approach and the characteristics of three office rent CPPIs are examined. Academics' and practitioners' 'consensus' models of office rent CPPIs are formed from a literature review and a survey, respectively. These models are compared in terms of their ex post predictive capability and it is found that generally, the better fitting models do not provide better predictions and in nearly 45% of cases choice of a model according to its fit will provide a worse model than a selection made at random. The academics' model generally predicts the Investment Property Databank (IPD) index better than the Jones Lang Wootton (JLW) and Investors' Chronicle Hillier Parker (HP) indices. It is generally the case that the one year ahead predictions are relatively poor.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available