The Behavioral Mechanism and Forecasting of Beijing Housing Prices from a Multiscale Perspective

Abstract

The first-hand house price in Beijing, the capital of China, has skyrocketed with 43 percent annual growth from 2005 to 2017, exerting tremendous adverse effects on people’s livelihood and the development of real estate. Thus, exploring the behavioral mechanism and accurate forecasts of house prices is a critical element in making decisions under uncertain conditions and is of great practical significance for both participants and policymakers in real estate. According to the complex features of house price, including nonlinear, nonstationary, and multiscale, and considering the remarkable time and frequency discrimination capability of multiscale analysis in dealing with house price problems, we develop an ensemble empirical mode decomposition- (EEMD-) based multiscale analysis paradigm to investigate the behavioral mechanism and then obtain accurate forecasts of house prices. Specifically, the monthly house price in Beijing over the period January 2005 to November 2018 is first decomposed into several different time-scale intrinsic-mode functions (IMFs) and a residual via EEMD, revealing some interesting characteristics in house price volatility. Then, we compose the IMFs and residual into three components caused by normal market disequilibrium, extreme events, and the economic environment using the fine-to-coarse reconstruction algorithm. Finally, we propose an improved hybrid prediction model for forecasting house prices. Our experimental results show that the proposed multiscale analysis paradigm is able to clearly reveal the behavioral mechanism hidden in the original house price. More importantly, the mean absolute percentage errors (MAPEs) of the proposed EEMD-based hybrid approach are 5.62%, 7.24%, and 8.63% for one-, three-, and six-step-ahead prediction, respectively, consistently lower than the MAPE of the three competitors.

Rent index forecasting through neural networks

“…Geographically, machine learning modeling has been applied for price-related forecasts in housing markets across many different countries and/or regions, including the USA (Nghiep and Al, 2001; Peterson and Flanagan, 2009; Park and Bae, 2015; Plakandaras et al. , 2015), China (Gu et al , 2011; Ho et al , 2021; Lam et al , 2008; Li et al , 2020; Liu and Liu, 2019; Xin and Runeson, 2004; Xu and Li, 2021), Spain (Rico-Juan and de La Paz, 2021), Italy (Morano and Tajani, 2013; Chiarazzo et al. , 2014; Morano et al.…”

“…the gross national product, gross domestic product, stock market index, consumer price index, default rate, interest rate and unemployment) for valuations (Kang et al. , 2020), from prices themselves for technical forecasts (Gu et al , 2011; Li et al , 2020; Xin and Runeson, 2004), from house-related characteristics for technical forecasts (Chen et al , 2017; Embaye et al , 2021; Igbinosa, 2011; Kang et al , 2020; Kitapci et al , 2017; Lam et al , 2008; Liu and Liu, 2019; Morano and Tajani, 2013; Nghiep and Al, 2001; Park and Bae, 2015; Rico-Juan and de La Paz, 2021; Terregrossa and Ibadi, 2021; Yasnitsky et al , 2021) and from macroeconomics for technical forecasts (Azadeh et al , 2014; Kang et al , 2020; Lam et al , 2008; Liu and Liu, 2019; Plakandaras et al , 2015; Rico-Juan and de La Paz, 2021; Xin and Runeson, 2004; Yasnitsky et al , 2021).…”

“…Some previous studies focus on using a specific machine learning model for price-related forecasting in the housing market (Abidoye and Chan, 2017; Azadeh et al , 2014; Chen et al , 2017; Chiarazzo et al , 2014; Gu et al , 2011; Igbinosa, 2011; Kang et al , 2020; Kitapci et al , 2017; Lam et al , 2008; Li et al , 2020; Morano et al , 2015; Xin and Runeson, 2004; Yasnitsky et al , 2021), some on making comparisons among different machine learning models (Park and Bae, 2015; Li et al. , 2017; Liu and Liu, 2019; Ho et al.…”

“…, 2015; Liu and Liu, 2019), between 4 and 5% (Kang et al. , 2020), between 5 and 6% (Kang et al , 2020; Li et al , 2020; Plakandaras et al , 2015), between 6 and 7% (Kang et al. , 2020), between 7 and 8% (Kang et al.…”

“…To authors' knowledge, this is the first study specifically on rent index in Chinese housing market. Analysis here should help expand knowledge of using machine learning models to forecast rental prices in the substantially growing Chinese market because previous research generally focuses on one certain city (Ho et al , 2021; Lam et al , 2008; Li et al , 2020; Liu and Liu, 2019; Xin and Runeson, 2004). The analysis should also help shed light on more recent evidence of usefulness of machine learning to forecast house price related series in China because previous research examines earlier time periods of 1981Q1–2002Q4 (Xin and Runeson, 2004), December 2010–October 2017 (Liu and Liu, 2019), January 2005–November 2018 (Li et al , 2020) and June 1996–August 2014 (Ho et al.…”