Search

The property valuation is classically challenged by the weighting of attributes and observed prices of properties. Modern approaches, using statistical potential, optimization, and rich data increased expectations for better results. Although the practitioners recognize that, they prefer accessible estimations with the convenience of interpreting them like in the sales comparison adjustment-grid. The weighting process is thus either weak in practice or complex in modeling culture. The challenge mainly originates from the incoherencies, sometimes significant, between the negotiated prices and the utility reward (contribution) of attributes. Economic agents take decisions under uncertainty, essentially influenced by their subjective expectations, capacity of evaluation and ability of using reliable information. Therefore, existing approaches suffer from properly addressing these issues of valuation. In this paper, we are motivated to improve these issues based on Choquet integral (CI) and fuzzy logic, well-known in Multiple-criteria decision-making (MCDM) problems. In comparison to the available classic (Adjustment-grid) and statistical approaches (like OLS regressions), the method we develop objectively and accurately weights the contribution of property attributes and the adjusted prices in market value estimations. Based on empirical experiments in case-by-case and mass valuation of single-family properties in Montreal (Canada), the results are encouraging and support its reliability. [ABSTRACT FROM AUTHOR]