Refinement of Storey enclosure forecasting method

Franco K.T. Cheung*, Martin Skitmore

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

2 Downloads (Pure)

Abstract

Previous researches have suggested that there is only little improvement in the accuracy of building forecasts as design develops. It has been criticized that established conventional forecasting methods also lack measures of their own performance. An early stage price forecasting model, the Storey Enclosure Method developed by James in 1954, uses some physical measurements of buildings to estimate building prices. Although James’ Storey Enclosure Method (JSEM) is not a widely used model in practice, the model has been proved empirically, if rather crudely, to be a better model than other commonly used models. This paper describes some preliminary research to refine JSEM using regression techniques. Advanced features of the proposed model include the use of cross validation for reliability analysis that simulates how forecasts are produced in practice and a dual stepwise selection strategy that enhances the chance of identifying the best model. To precisely judge the performance of models, this paper suggests using bias and consistency with parametric and nonparametric statistical inferences.
Original languageEnglish
Title of host publication2005 AACE International Transactions: AACE International's 49th Annual Meeting
EditorsM. Gelhausen
PublisherAACE International
Pages1-5
Number of pages5
Publication statusPublished - 2005
Externally publishedYes
Event2005 AACE International 49th Annual Meeting - New Orleans, LA, United States
Duration: 26 Jun 200529 Jun 2005

Publication series

NameAACE International Transactions
ISSN (Print)1528-7106

Conference

Conference2005 AACE International 49th Annual Meeting
CountryUnited States
CityNew Orleans, LA
Period26/06/0529/06/05

Fingerprint

Dive into the research topics of 'Refinement of Storey enclosure forecasting method'. Together they form a unique fingerprint.

Cite this