House Price Index Research Articles

Data‐sparse price indexes by spatio‐temporal regularization and PCA: An application to the Australian housing market

AbstractWe present two novel approaches to overcome the limitations of data‐sparse local house price indexes and combine them into a single model pipeline that is simple, computationally efficient, and interpretable. The first contribution is a new spatio‐temporal regularization of least squares dummy variable models, such as the repeat sales regression used here. This regularization encodes prior knowledge of the proximity of houses in space and their sales in time. It handles missing values in a natural way. The second is nonlocal regularization using truncated principal component analysis (PCA) applied to the resulting national collection of local price indexes. The PCA loadings show that there are important underlying socioeconomic factors that can be leveraged in the construction of Australian market indexes. This PCA reveals important socioeconomic factors, showing that many local markets can be described by a few broad aspects of the national market, consisting of a general trend that contrasts regions influenced by the mining industry with Sydney and Melbourne, and another trend that highlights lifestyle.

Does US Dollar Supply Trigger House Price Fluctuations? An International Comparison

Cyclical fluctuations in asset prices are closely related to the overall stability of the macroeconomy and have thus become a significant topic in academic and professional exploration. Among these fluctuations, changes in the supply of the U.S. dollar, the global reserve, and primary settlement currency have a broad and far-reaching impact on international financial markets. Since COVID-19, asset prices across various countries have shown differing trends due to fluctuations in the U.S. dollar supply. Housing prices, a key focus in asset pricing research, have been affected to varying degrees. This thesis utilizes seven house price indices from representative cities worldwide from 2020 to 2023 to compare the differential impacts across various countries using an empirical approach. The results indicate that the impact of the U.S. dollar money supply (M2) diminishes in a gradient from the U.S. to East Asia, and then to Europe.

An approach for constructing spatially paired pseudo repeat-sales housing price indices in China

AbstractConstructing a reliable housing price index is crucial for accurately reflecting housing price dynamics and enhancing transparency in the housing market. However, existing methods for constructing housing price indices often confront many challenges, such as omitted variable issues in the hedonic model and constrained samples used in the repeat-sales model. To address these challenges, this study identifies a spatially paired relationship between complexes and develops a Spatially Paired Pseudo Repeat-Sales model to construct the housing price index. This approach offers two significant advantages: first, it enlarges the sample size used in the repeat-sales model at least 1.8 times; second, it effectively estimates the effects of spatial dependency and physical housing factors on prices while mitigating the impacts of unobservable factors through differentiation. The findings of this paper suggest that using the spatially paired pseudo repeat-sales model can significantly improve the estimation of housing price volatility, by approximately 13%. Moreover, the constructed housing price index model demonstrates significant robustness, even when the spatial weight settings are altered. This research provides convincing evidence of improved housing market transparency for stakeholders, including governments, institutional developers, and individual investors. Additionally, the constructed model can inform other similar research in housing price index construction by providing a spatially pairing perspective on housing complexes.

The Impact of Taxation on the Cost of Living: A Comprehensive Analysis

This study examines the taxation-related factors affecting Malaysia's cost of living. The rising cost of living, a persistent issue in economics and social well-being, is often linked to global challenges faced by emerging nations. Direct tax has been identified as a primary contributor to the financial strain on households, who must manage various essentials beyond food, clothing, and shelter, such as education, cellular services, and transportation. The study analyzes the relationship between seven independent variables—direct tax, unemployment rate, exchange rate, household consumption expenditure, house price index, subsidies, and net trade—and the cost of living in Malaysia, using 30 years of annual data (1992-2022) with EViews for econometric analysis. This study uses descriptive statistical analysis for the description of data, coefficient matrix analysis and regression analysis for determining the impact of dependent and independent variables. The findings show that the cost-of-living pressures are positively correlated with direct tax, housing price index, and net trade. Data was sourced from the National Property Information Centre (NAPIC), the World Bank, and the Economic Planning Unit (EPU). The analysis underscores the significant impact of direct tax on the cost of living. To mitigate this impact, governments could implement fair and progressive tax reforms, provide targeted subsidies, and promote economic growth to boost income levels. Additionally, social programs and efficient government investment can further reduce household financial burdens.

Performance evaluation of housing price index prediction model using DNN and SVR

This study used the base interest rate, mortgage interest rate, deposit interest rate, consumer price index, exchange rate, and composite stock price index as variables to predict the housing price index. The variables were normalized using the min-max normalization method. Machine learning methods, Deep Neural Networks (DNN) and Support Vector Regression (SVR), were employed to construct a prediction model, and the fit of the model was evaluated using the Root Mean Square Error (RMSE). The analysis results of the DNN model indicated that when applying the softplus activation function, the RMSE was 0.1298, resulting in the smallest prediction error, and the correlation coefficient between the predicted and actual values was highest at 0.9342. The analysis results of the SVR model using the Gaussian RBF kernel showed that the RMSE was smallest at 0.2781 when the parameters were set to C=100 and Gamma=0.1. Therefore, the DNN model was found to be more fit for predicting the housing price index. Additionally, it was confirmed that the consumer price index had a non-linear effect with a positive relationship, and the base interest rate had a linear effect with a negative relationship.

USE OF MACHINE LEARNING AND DEEP LEARNING METHODS IN HOUSING PRICE INDEX ESTIMATION: AN ANALYSIS ON ANKARA AND ISTANBUL

Factors such as supply chain difficulties, rising energy and oil prices, economic recession and production loss due to the pandemic have increased costs and inflation. All these factors have also seriously affected the construction sector. This study aims to create a deep learning and machine learning focused forecasting system based on Istanbul and Ankara monthly housing price index data for the period of January 2010 to June 2023. The system was created using approximately 13 years of housing interest rates, Consumer Price Index, XGMYO, Monthly Average Dollar and XAU data as the basis of the Istanbul and Ankara Housing Price Index forecasting process. During the research process, different RNN structures (Long and Short Term Memory, Gated Recurrent Unit) and machine learning (Random Forest) structures were tested and the effectiveness of these structures in housing price index forecasting was compared. The performances of the models were evaluated using RMSE, MSE, MAE, MAPE and R2 statistics. According to the results obtained, the method that gave the best performance for both provinces is the RF model. This is followed by LSTM and GRU models, respectively

Structural analysis of reverse mortgages in Taiwan

This study investigated the framework of reverse mortgage (RM) programs in Taiwan, with a particular emphasis on the perspectives of both debtors and creditors. Employing the Lee–Carter model, the research estimated mortality rates in Taiwan while utilizing the theoretical framework to analyze various aspects of RM. House prices were assessed using the house price index as a foundational metric for estimating prices, along with lump-sum loanable amounts and RM considerations. Empirical analysis was conducted on debtors’ loans and incomes and financial institutions’ perspectives. The study delved into parameters such as maximum loanable amounts and income replacement ratios for debtors based on demographic factors such as sex, age, and occupation. Furthermore, the structure of RM programs was scrutinized from the viewpoint of financial institutions, considering factors such as rental income, remaining house value, creditor insurance protection, premium expenses, and creditor profitability. A series of simulations were performed to gauge the effects of different variables on the outcomes. The normal distribution of the house price index (HPI), as verified through the Anderson-Darling (AD) test and Kolmogorov–Smirnov (KS) test, bolstered the theoretical findings. This study’s primary contribution lies in its comprehensive exploration of the structure and advantages of RMs from the dual perspectives of debtors and creditors, a departure from insights from previous research predominantly focused on financial institutions. This research can aid debtors and creditors in evaluating RM programs and making informed decisions regarding asset allocation, thereby facilitating the selection of more suitable programs tailored to debtor characteristics.

Differential Measurement Error in House Price Indices

Differential Measurement Error in House Price Indices

A stochastic loss given default model of Hungarian residential mortgages

AbstractThe paper analyses the Loss Given Default (LGD) rates of residential mortgages, using a model based on stochastic collateral value. The implementation of the model is based on exponential Ornstein-Uhlenbeck processes fitted to the Hungarian regions' house price indices. According to the model results, in case of a mortgage with a 80% loan-to-value ratio at origination, the expected LGD is around 30–40%, depending on the region. The highest LGD rates are estimated for villages, while the lowest rates are expected in Budapest and cities in the middle of the country. The range of the regional differences can reach 7 percentage points. According to the LGD Risk index based on the aggregated model, the LGD risk profile of recently issued mortgages has improved significantly since 2009 in Hungary.Due to the strong negative relation between the house prices and mortgage default rates, the expected return on defaulted collateral value tends to be low. The results could be relevant for credit institutions in their mortgage origination decisions and enhance analysis of lending processes and the associated risks.

Application of the ARIMA model in house price index in Malaysia

The factors that affecting the escalating price of houses in Malaysia are driven by factors such as population growth, income dynamics, interest rates, and GDP. This phenomenon has notably outpaced the growth of household incomes, thus majorly impacting Malaysians. The study’s primary goal is to forecast the housing price index in Malaysia from the best model obtained using Box-Jenkins method. aligning with the 2018-2025 National Housing Policy objectives, utilizing advanced machine learning and time series modeling. The objectives guide the research: to find the best model for predicting house price index in Malaysia using Box-Jenkins Method. Utilizing secondary data from the National Property Information Centre (NAPIC) spanning from 1988 to 2023, the study employed the analytical method of ARIMA. The results favored the ARIMA (1,1,1) model as the best model in predicting housing price indexes. This offers an excellent forecasting model for residential properties towards gaining better understanding of their pricing dynamics and offers potential solutions to the issue of housing affordability for Malaysians.

Housing Market Appreciation and the White-Black Wealth Gap

Abstract Real house prices in the United States have risen by 55 percent over the last four decades, driving substantial wealth benefits to homeowners. However, research has not explored how this rise in house prices has affected White-Black wealth gaps, or the mechanisms that may underlie this relationship. Using geocoded longitudinal household-level wealth data from the Panel Study of Income Dynamics and tract-level house price index data, I estimate that housing market appreciation between 1984 and 2021 explains 70 percent of the increase in the median White-Black wealth gap over this period. I find that most of this effect is due to White-Black gaps in homeownership, with White-Black gaps in house values playing a smaller role. In contrast to recent findings about racialized housing markets, I do not find that gaps in neighborhood house price appreciation between White and Black homeowners contributed to White-Black wealth gaps in the 2000s and 2010s. These results highlight the importance of cumulative advantage processes in driving wealth inequalities and demonstrate how the legacies of institutional racism contribute to contemporary racial wealth gaps.

Pre-owned housing price index forecasts using Gaussian process regressions

Purpose The purpose of this study is to make property price forecasts for the Chinese housing market that has grown rapidly in the last 10 years, which is an important concern for both government and investors. Design/methodology/approach This study examines Gaussian process regressions with different kernels and basis functions for monthly pre-owned housing price index estimates for ten major Chinese cities from March 2012 to May 2020. The authors do this by using Bayesian optimizations and cross-validation. Findings The ten price indices from June 2019 to May 2020 are accurately predicted out-of-sample by the established models, which have relative root mean square errors ranging from 0.0458% to 0.3035% and correlation coefficients ranging from 93.9160% to 99.9653%. Originality/value The results might be applied separately or in conjunction with other forecasts to develop hypotheses regarding the patterns in the pre-owned residential real estate price index and conduct further policy research.

A Regional Analysis of the Impacts of Changes in Housing Values on Property Tax Assessment During and After the 2008 Financial Crisis

This study examines the relationship between changing house prices and assessed value in the Detroit Michigan region. Using parcel-level data from 2008 through 2016, we construct a Hedonic Housing Price Index and an Assessed Value Index for 120 Michigan assessing jurisdictions in the five-county region surrounding Detroit (Macomb, Monroe, Oakland, Washtenaw, and Wayne). Our findings are consistent with the literature, indicating at least a three-year lag between changes in house prices and changes in property assessments. Communities considered to be distressed tended to over-assess properties before the crisis, but these communities recalibrated assessments to better align with market conditions after the financial crisis. Our evaluation also shows that while assessed values decreased more slowly than housing prices fell during the crisis, assessed values tended to fall to even deeper levels than housing prices over a longer time horizon. JEL Codes: H71, H83, G01

주택매매시장에서의 거품에 관한 실증연구

This study examines the monthly real house price index movements over the period 1986-2023 using the GSADF method to determine whether bubbles existed in the Korean housing market, where they occurred, when they occurred, and how long they lasted by housing type (total, apartment, detached house, and townhouse). We summarise the main findings from our empirical analysis as follows. First, there were multiple bubbles in housing prices across the nationwide, Seoul, and all provincial metropolitan areas. However, there are many differences in how bubbles occurred across regions and types of housing. Second, the post-bubble period lasted much longer in the detached house market than in the apartment and row house markets. Third, many housing market bubbles occurred during periods of abundant liquidity. Fourth, while most bubbles are positive, appearing during periods of rapid house price appreciation, negative bubbles appeared in some areas (Busan, Gwangju, and Ulsan) during the 1998 Asian financial crisis. Fifth, the detached house markets in Busan and Ulsan are found to be bubbling at the end of the sample (December 2023). These results suggest that the GSADF test can not only detect historical bubble periods, but can also detect bubbles in real time, providing early warning of bubbles.

What drives the real estate market? Could behavioral indicators be useful in house pricing models?

PurposeThe objective of this paper is to examine the determinants of the Greek house market during the period 2006–2022 using not only economic variables but also behavioral variables, taking advantage of available information on the volume of Google searches. In order to quantify the behavioral variables, we implement a Python code using the Pytrends 4.9.2 library.Design/methodology/approachIn our study, we assert that models relying solely on economic variables, such as GDP growth, mortgage interest rates and inflation, may lack precision compared to those that integrate behavioral indicators. Recognizing the importance of behavioral insights, we incorporate Google Trends data as a key behavioral indicator, aiming to enhance our understanding of market dynamics by capturing online interest in Greek real estate through searches related to house prices, sales and related topics. To quantify our behavioral indicators, we utilize a Python code leveraging Pytrends, enabling us to extract relevant queries for global and local searches. We employ the EGARCH(1,1) model on the Greek house price index, testing several macroeconomic variables alongside our Google Trends indexes to explain housing returns.FindingsOur findings show that in some cases the relationship between economic variables, such as inflation and mortgage rates, and house prices is not always consistent with the theory because we should highlight the special conditions of the examined country. The country of our sample, Greece, presents the special case of a country with severe sovereign debt issues, which at the same time has the privilege to have a strong currency and the support and the obligations of being an EU/EMU member.Practical implicationsThe results suggest that Google Trends can be a valuable tool for academics and practitioners in order to understand what drives house prices. However, further research should be carried out on this topic, for example, causality relationships, to gain deeper insight into the possibilities and limitations of using such tools in analyzing housing market trends.Originality/valueThis is the first paper, to the best of our knowledge, that examines the benefits of Google Trends in studying the Greek house market.

Persistence analysis of the real estate US index and major cities

This paper analyses the stochastic properties of the US real estate market in the long term (1987–2023). For this purpose, we rely on fractionally integrated methods to fill the gap left by studies focusing on possible cointegration that use linear unit root tests. Both original and log transformed data of house price indices have been used with and without taking inflation into account. Even though several studies mentioned mean reversion in the short term, in the long term the real estate market shows a clear pattern of high persistence and no reversion to the mean. The effect of inflation seems to add correlation to real prices and conversely, an increase in the integration factor. Structural breaks reveal that recent values of this integration factor are greater, thus prices are becoming more persistent.

Spatial and temporal diffusion of property prices and sales across British regions

Purpose The purpose of this study is to reveal the dynamics of house prices and sales in spatial and temporal dimensions across British regions. Design/methodology/approach This paper incorporates two empirical approaches to describe the behaviour of property prices across British regions. The models are applied to two different data sets. The first empirical approach is to apply the price diffusion model proposed by Holly et al. (2011) to the UK house price index data set. The second empirical approach is to apply a bivariate global vector autoregression model without a time trend to house prices and transaction volumes retrieved from the nationwide building society. Findings Identifying shocks to London house prices in the GVAR model, based on the generalized impulse response functions framework, I find some heterogeneity in responses to house price changes; for example, South East England responds stronger than the remaining provincial regions. The main pattern detected in responses and characteristic for each region is the fairly rapid fading of the shock. The spatial-temporal diffusion model demonstrates the presence of a ripple effect: a shock emanating from London is dispersed contemporaneously and spatially to other regions, affecting prices in nondominant regions with a delay. Originality/value The main contribution of this work is the betterment in understanding how house price changes move across regions and time within a UK context.

Herding and reverse herding in US housing markets: new evidence from a metropolitan-level analysis

ABSTRACT This study is the first to examine herding and reverse herding in US metropolitan housing markets based on Zillow ZIP-level house price indices. Reverse herding is found to be more prevalent than herding, which differs markedly from equity markets and outcomes derived from less granular house price indices. The results suggest that the interaction between price appreciation and overconfidence may drive reverse herding. Also, herding and reverse herding show strong dependency on market conditions. Wide spatial and temporal variation in herding and reverse herding suggests the importance of local characteristics as determinants of the rationality of market responses.

VECM 모형을 이용한 임차권등기명령에 영향을 미치는 요인 분석

This study analyzed factors influencing the non-returned of Chunsei deposits quantitatively by using the number of lease registration orders as the dependent variable and corporate bond interest rates, Chunsei rate, Chunsei price index, and index of actual transaction prices for residential property as independent variables, using vector error correction model. The shock response analysis showed that corporate bond interest rates and Chunsei rate exhibited a positive effect on lease registration orders, while the Chunsei price index and the index of actual transaction prices showed a negative effect. The variance decomposition analysis revealed that initially, the Chunsei price index had a significant influence on the occurrence of lease registration orders, while corporate bond interest rates consistently exerted a significant influence at all time lags. The implications drawn from this study suggest that for preventing incidents such as non-returned of Chunsei deposits, the government needs to thoroughly analyze and continuously monitor changes in interest rates, Chunsei price index, and House price index, and formulate and implement policies tailored to the market conditions.

Predictive strength of Macroeconomic Imbalance Procedure indicators

Motivation: The Macroeconomic Imbalance Procedure (MIP) is a key step in the European Semester, aimed at the coordination of the economic policies of the EU Member States to prevent excessive macroeconomic imbalances in the EU and support structural reforms. The MIP was originally envisaged as a legal tool for crisis prevention, allowing macroeconomic imbalances to be detected and then remedied, but is also used as an Early Warning System. However, the real strength of MIP indicators to predict crises has not been proved in practice and is widely contested in the literature.Aim: Fourteen scoreboard (“main”) and 28 auxiliary MIP indicators are currently in use. This paper is aimed at the assessment of the power of all MIP indicators in predicting crises.Results: The added value of our research is to test the MIP’s ability to predict changes in GDP, which may be considered as a proxy for the deterioration or improvement of the economic situation. Very little investigation has been done in this area so far. In addition, to our knowledge, no research papers have investigated the relevance of auxiliary MIP indicators. Our results show that only four main indicators (house price index, nominal unit labour cost index, general government sector debt, and export market shares) and another four auxiliary indicators (residential construction as percentage of GDP, activity rate, people living in households with very low work intensity, and export performance against advanced economies) seem to be able to predict the upcoming crises.