Universiti Teknologi Malaysia Institutional Repository

Spatial Autocorrelation in hedonic model: Empirical Evidence from Malaysia

Ismail, Suriatini and Mar Iman, Abdul Hamid and Mohd Ali, Hishamuddin and Kamaruddin, Norhaya and Sipan, Ibrahim @ Atan and Navaneethan, Renuga (2008) Spatial Autocorrelation in hedonic model: Empirical Evidence from Malaysia. In: International Real Estate Research Symposium (IRERS) : Benchmarking, Innovating and Sustaining Real Estate Market Dynamics (2008 : Kuala Lumpur, Malaysia) , 28 – 30th April 2008, PWTC Kuala Lumpur, Malaysia. (Unpublished)

[img]
Preview
PDF
221Kb

Abstract

The need to consider spatial autocorrelation in hedonic price modelling is paramount since reliability and accuracy are important in any housing market analysis. Although hedonic modelling is widely accepted, the issue of spatial autocorrelation has received little attention in house price studies in Malaysia. This paper examines the presence of spatial autocorrelation in the hedonic price model using data from Malaysia. A pilot sample of eighty three-year sale observations (2001-2003) and nine property attributes were used to construct four types of double-log competing models namely ordinary least square model (OLS), spatial autoregressive model (SAR), spatial error model (SEM) and general spatial model (SAC). The results have indicated the presence of spatial lag and spatial error types of autocorrelation based on the SAC model. However, based on the out-sample test, the models were found to be equivalent in terms of statistical quality. Despite the findings, it is important to stress that taking into account the possible phenomenon of spatial autocorrelation is critical in using hedonic housing price model to ascertain that the model chosen is free from spatial autocorrelation problems. If the problem does exist, we can adopt the appropriate steps to improve the statistical quality of price-estimating model being chosen.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:hedonic model, house price, spatial autocorrelation
Subjects:G Geography. Anthropology. Recreation > G Geography (General)
T Technology > T Technology (General)
G Geography. Anthropology. Recreation > GA Mathematical geography. Cartography
H Social Sciences > H Social Sciences (General)
H Social Sciences > HG Finance
H Social Sciences > HA Statistics
Divisions:Geoinformation Science And Engineering (Formerly known)
ID Code:5584
Deposited By: Dr Suriatini Ismail
Deposited On:23 May 2008 02:25
Last Modified:01 Jun 2010 15:32

Repository Staff Only: item control page