![Sample our Built Environment journals, sign in here to start your access, Latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5926/TOC-Subject-Sample-2021-Built-Environment.jpg"})
Abstract
The primary aim of this paper is to introduce valuation modeling applicable to thin housing markets, with a focus on the Arab housing sector in Israel. The estimation procedure utilizes two input values: transaction data and subjective valuations provided by property owners, the data for which are derived from the Israel Tax Authority (ITA) and the Household Expenditure Survey (HES). Average property values are also weighted and ranked according to location, size, and average income factors. The main contribution of these modeling techniques is that they can be employed to estimate the residential property values in markets that experience a low frequency of housing transactions and where information is limited, with the added benefit of understanding housing value movement and market dynamics. Housing policies could be influenced by this deeper understanding of house price behavior within localities and submarkets, potentially with the ability to monitor changes in dwelling values and segmentation and segregation effects.
Notes
1 The exchange rate of NIS to USD was 3.58 in 2011.
2 Critically, if sufficient data relating to property values were available, a cross-validation (CV) method could be applied using the learning sample (say, 20% of observations) upon which the prediction values and accuracy indices were calculated. However, in this instance we have no reliable 'ground truth' available. Therefore, this method for testing the accuracy of results for assessment purposes cannot be used here. Instead, we employ an alternative approach to estimate and validate the latent continuous variables applying a weighing procedure.