Developing a Demand Forecasting Model for Offices in Kuala Lumpur

 DEVELOPING A DEMAND FORECASTING MODEL FOR OFFICES IN KUALA LUMPUR

Project Leader

Prof Dr Md Nasir bin Daud

Project Members

Assoc. Prof. Dr. Sr Rosli Said

Assoc. Prof. Dr. Sr Anuar Alias

Dr. Sr Yasmin Mohd Adnan

Sr Zulkifli Esha

Sr Zahiriah Yahya

Prof. Dr. Rasimah Arifin

Research Assistant

Suhaila Suhaimi

ABSTRACT

The aim of this research has been to develop a viable econometric model for forecasting office demand in Kuala Lumpur. A primary motivation was the recognition that it is desirable to generate the capability to forecast office space demand so as to facilitate the planning of its future supply, as is also alluded to by the Kuala Lumpur Structure Plan 2020. There has been a visible drift in the supply of new and quality office accommodations away from the old CBD area of Jalan Tuanku Abdul Rahman toward the new centre of gravitation of the Golden Triangle, particularly around the iconic Petronas Twin Towers, attributable to the changing patterns of supply and demand for office accommodations the interplays of which lead to varying outcomes in different locations.

This study's effort to construct an office demand forecasting model for Kuala Lumpur was based on experimentation with an original model developed by the Royal Institution of Chartered Surveyors (RICS) for the City of London. In essence, the approach in use had relied on Structural Econometric Modelling technique. Modifications were made to the London model in order to suit the local conditions of Kuala Lumpur, resulting in the Introduction of 7 interlinked equations and 1 identity into the structural modelling ecosystem on the basis of & endogenous and 5 exogenous variables used. Trial runs were then undertaken on the datasets collected for Kuala Lumpur. Unfortunately modelling could not go much further because of flaws in the datasets, particularly the data on stock withdrawal. This had prevented a viable model from being generated.

Further effort must continue in order to pursue the aim of a forecasting model, but certain other approaches must now be considered that dispense with the need to rely on the same datasets. This can come as an extension to the current project.