Use this URL to cite or link to this record in EThOS:
Title: Automated mass appraisal system with cross-city evaluation capability : a test development in China
Author: Guo, Yihan
ISNI:       0000 0004 8505 0647
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
The appraisal of property value is extremely important in a modern economy. For example, developers and end-consumers use appraisals for their investment decisions. Governments use it for taxation purposes, while banks rely on appraisals to update their risk profile when managing mortgage and credit application activities. With fast developing economies, quickly valuing new cities and suburbs as they get built becomes particularly difficult. Globalisation has also increased the need for common international valuation standards and automated methods. This research investigates the present mass appraisal systems and the role of automated valuation models. Financial institutions and institutional investors are increasingly more concerned about constantly updating their present portfolio value especially in a dynamic market. Trends of significant peaks and troughs need to be accounted in a faster cycle time with short bursts of pricing adjustments. The problem poses a challenge because property transactions are infrequently traded unlike other commodities such as securities. Hence, there are not many recent transactions for the same property to receive an updated value with a simple adjustment based on economic conditions. The study proposes a method that solves both large-scale mass appraisal with an ability to search across cities to discover properties with similar characteristics for its update and comparison scheme. This research advances the automated valuation model for the residential property market with a test development performed in China. In particular, the resulting model was tested with data from Chinese Tier 1, 2 and 3 Cities to evaluate property values. This research performs several major accomplishments. First, it demonstrates the efficacy of reducing human cognitive effort in the mass appraisal exercise. Second, by applying Artificial Neural Network capabilities in the automated valuation model, pricing of residential properties are able to draw upon knowledge from more mature cities with greater number of transactions and apply to newer developments in less developed cities. Third, the proposed mass appraisal system shows the reliability and robustness that matches the rapid development of Chinas real estate market that had been verified by a real application. Finally, the approach developed provides a valuable new method for property valuation that reduces the possible bias, increases consistency and lowers the effort required by current manual methods, with a lower data requirement.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: HD Industries. Land use. Labor