A two-stage variational jump point detection algorithm for real estate analysis

Real estate is one of the most important forms of portfolio, which plays an essential role in the economy of Hong Kong. Extensive investigations indicate that jump points in the private domestic price indices are subject to various factors such as international events, regional incidents, and government policies. Detecting jump points and identifying their relationships with these factors are indispensable since such findings can facilitate policymakers and investors to look forward. In this paper, we propose a novel two-stage variational methodology using the total variation minimization algorithm and derivative-based detector with the thresholding function for detecting jump points in the Hong Kong private domestic price indices. Our algorithm can weaken statistical fluctuation of the price indices via the total variation formulation and accurately detect jump points based on its rate of change. Comparative experimental results based on the monthly private domestic price indices from January 2010 to July 2020 reveal that the proposed method can detect 40 jump points corresponding to the aforementioned factors whereas the current state-of-the-art empirical mode decomposition approach can only detect 29 jump points, which indicates the superior performance of the proposed method.