House Price Index Research Articles

기계학습 방법론을 활용한 아파트 매매가격지수 연구*

This study proposes to calculate new housing price indices through machine learning techniques. Our research is conducted focusing on the random forest and artificial neural network methodologies that proved excellence in existing real estate studies. We use micro-level real estate transaction data and housing characteristics information to train our models. As a result, the forecasting powers of the machine learning based models are found to be much superior in terms of explanatory powers and estimation performances compared to the hedonic methodology- based model. And the random forest model shows the best explanatory power and performance. Our results show that the housing price indices based on the machine learning models have greater volatility than currently used indices at the time of the housing price increase. Considering the limitation that the existing indices have a smoothing problem, our results can be interpreted that the new machine learning based indices reflect the market trend successfully.

Robust multiscale estimation of time-average variance for time series segmentation

There exist several methods developed for the canonical change point problem of detecting multiple mean shifts, which search for changes over sections of the data at multiple scales. In such methods, estimation of the noise level is often required in order to distinguish genuine changes from random fluctuations due to the noise. When serial dependence is present, using a single estimator of the noise level may not be appropriate. Instead, it is proposed to adopt a scale-dependent time-average variance constant that depends on the length of the data section in consideration, to gauge the level of the noise therein. Accordingly, an estimator that is robust to the presence of multiple mean shifts is developed. The consistency of the proposed estimator is shown under general assumptions permitting heavy-tailedness, and its use with two widely adopted data segmentation algorithms, the moving sum and the wild binary segmentation procedures, is discussed. The performance of the proposed estimator is illustrated through extensive simulation studies and on applications to the house price index and air quality data sets.

Micro-geographic property price and rent indices

We develop a programming algorithm that predicts a balanced-panel mix-adjusted house price index for arbitrary spatial units from repeated cross-sections of geocoded micro data. The algorithm combines parametric and non-parametric estimation techniques to provide a tight local fit where the underlying micro data are abundant, and reliable extrapolations where data are sparse. To illustrate the functionality, we generate a panel of German property prices and rents that is unprecedented in its spatial coverage and detail. This novel data set uncovers a battery of stylized facts that motivate further research, e.g. on the positive correlation between density and price-to-rent ratios in levels and trends, both within and between cities. Our method lends itself to the creation of comparable neighborhood-level rent indices (Mietspiegel) across Germany.

Housing price indices from online listing data: Addressing the spatial bias with sampling weights

This paper proposes a method to reduce the inherent sampling bias when estimating housing price indices using online listing data. Producing more accurate and representative metrics is important as new sources of data emerge with higher frequency, detail, and volume, providing more information for policymaking, but usually come with strong sampling biases that are often overlooked. In the case of housing price indices, although the literature around its estimation is abundant, it has concentrated only in traditional and formal sources of housing data, which is normally not available in some markets (i.e. renting) and locations (developing countries). In this paper, I propose a method to create a housing price index (HPI) that is comparable in quality to the industry-standard Case-Shiller HPI but using online listing data. Using online listing data from a developing economy (Chile), this paper shows that large sampling biases present when using raw unweighted data, how these biases can be minimized using sampling weights, and how new and relevant information can be obtained from adjusted HPIs that can lead better policymaking.

Urban land valuation with bundled good and land residual assumptions

This paper develops a new approach to estimate the value of urban land. We extend AMM theory by adding the assumption of partial irreversibility. Bundled goods assumptions imply that land value with a structure can evolve differently than as-if vacant value, even in the first decades of structure life. We develop a hybrid model that nests bundled goods with land residual methods and we develop a new test of predictive accuracy. Granular house price indices produced by machine learning are used to estimate hybrid economic structure and land values.We fit hybrid models to Maricopa County assessor data on houses up to 25 years old during a bust and recovery period (2007–2018). Results support structure value that evolves at some fraction (<1.0) of property value: i.e., partial bundling. The hybrid model achieves substantially lower Coefficients of Dispersion (CODs) than the land residual method, and both methods easily beat standard AVMs that do not separate land and structure values.

Exploring market choices in the US ornamental horticulture industry

AbstractIn 2019, the US ornamental horticulture industry had a farm gate crop value of roughly $17 billion, contributed $348 billion in total output to the nation's economy, and created 2.32 million employment opportunities. However, as the industry matures and reaches a slow‐growing stage, interfirm rivalry may appear among grower firms for marketing channels and sales opportunities. Independent garden centers and retailers report facing increased competition from mass merchandisers. Market channel choices, as part of the competitive strategy, are alternative ways for ornamental production to reach consumers. However, marketing practices of this industry have not received adequate attention from researchers despite the economic significance of the industry. Combining four waves of the National Green Industry Survey from 2004 to 2019, this study provides a comprehensive analysis of factors influencing firms' market participation and market channel choices. Over the survey period, traditional wholesale market channels such as landscape services companies, rewholesalers, and single‐location garden centers have remained mainstream channels. The direct‐to‐consumer (DTC) channel showed the potential to gain significant market share. Thirty percent of the firms participated in DTC channel in 2018, and their sales represent about 22% of total sales by all survey firms. Our empirical results show firms' wholesale market participation and choices of market channels are largely influenced by plant types and firm perceptions (variables potentially affecting firms' assessment of the “size” of their market) and variables approximating the market size such as regional dummies and housing price index, confirming the impact of market size on entry decision. [EconLit Citations: C50, M31, L14].

The long-term relationship between the US housing market and HNWIs

PurposeThe authors analyze the long-term relationship between the US housing market and the concentration of high net worth individuals.Design/methodology/approachThe authors apply a cointegration and causality approach on a ten years database. The database includes by US state the density of high net worth individuals, the house price index, the homeownership rate, the number of housing units, and the privately-owned housing units authorized by building permits.FindingsThe results show a limited cointegration between the concentration of US wealthy households and the housing market. Thus, the authors can partly predict the fluctuations transmission from the housing market to wealthy households.Originality/valueThe authors suggest that our findings could be of interest to wealthy households, real estate professionals, and public authorities.

Multidimensional Comparative Analysis of the House Price Index in Europe Between 2021-2022 in the Aspect of Economic Security

The article presents a multidimensional comparative analysis of the impact of crude oil and liquefied petroleum gasses (LPG) prices on the housing price index in 28 European countries. Research shows that the COVID-19 pandemic and the attack of Russia on Ukraine have led to similar large fluctuations in crude oil and LPG prices. This, in turn, contributed to an increase in inflation and house price indices in the 28 analyzed European countries. Such an increase was driven by higher total construction costs, including higher prices for materials, energy and charges for building services. Housing prices in respective European countries vary. The study examines the dynamics of their changes and compares them with the use of ranking.

THE EFFECTS OF CHANGES IN EXCHANGE RATE ON FOREIGN HOUSİNG DEMAND: IN TURKEY

Çalışmamızın amacı reel döviz kurundaki değişimlerin yabancı konut talebi üzerindeki etkisinin araştırılmasıdır. Bu amaçla yabancılara yapılan konut satışlarına döviz kurunun etkisi incelenmek istenmiştir. Bu çalışmada ülke uyruklarına göre yabancılara yapılan konut satış rakamları 2015 yılı itibariyle yayınlandığı için 2015-2021 dönemini kapsayan ülke uyruklarına ait aylık yabancılara yapılan konut satış verileri kullanılmıştır. Bu doğrultuda yapısal kırılmalı birim kök ve eşbütünleşme testleri kullanılmıştır. Zaman serisi analizinin kullanıldığı modelde bağımlı değişken yabancılara konut satışı olup bağımsız değişkenler olarak reel efektif döviz kuru ve konut fiyat endeksi rakamları kullanılmıştır. Gregory-Hansen eşbütünleşme testi sonuçlarına göre yabancılara konut satışı ile reel efektif döviz kuru ve konut fiyat endeksi arasında uzun dönemli eşbütünleşme ilişkisinin varlığı görülmektedir. Eşbütünleşme ilişkisi bulunan değişkenler arasında uzun dönemli ilişkinin analizi yapısal kırılmaların kukla değişken olarak analize dahil edildiği FMOLS katsayı yöntemi ile araştırılmış ve reel efektif döviz kurundaki %1’lik azalış (TL’nin değer kaybetmesi) yabancı konut satışını %3.43 arttırdığı sonucu elde edilmiştir. Konut fiyat endeksi değişkeninin katsayısı ise istatistiksel olarak anlamsız çıkmıştır.

CAUSALITY RELATIONSHIP BETWEEN STOCK PRICES, EXCHANGE RATE AND HOUSE PRICE INDEX

Bu çalışmada, hisse senedi fiyatları, döviz kuru ve konut fiyatları arasındaki nedensellik ilişkisi Türkiye özelinde ampirik olarak analiz edilmiştir. 2013:M3 – 2022:M1 dönemine ait aylık veriler kullanılarak, BIST100 endeksi, BIST İnşaat endeksi, USD/TL kuru ve Konut Fiyat endeksi arasındaki nedensellik ilişkisi araştırılmıştır. Seçilen değişkenler arasındaki ilişki Granger Nedensellik testi ile analiz edilmiştir. Bulgulara göre, BIST100 endeksi ile konut fiyat endeksi arasında, BIST İnşaat endeksi ile konut fiyat endeksi ve dolar kuru ile konut fiyat endeksi arasında çift yönlü nedenselliğin varlığı tespit edilmiştir. Dolar kuru ile BIST100 endeksi arasında, dolar kuru ile İnşaat endeksi arasında, BIST100 endeksi ile İnşaat endeksi arasında nedensellik ilişkisine rastlanmamıştır. Konut fiyatlarını etkilemek isteyen politika yapıcıların dolar kuru ve hisse senedi fiyatlarını da göz önünde bulundurması gerektiği ifade edilebilir.

TÜRKİYE’DE KONUT FİYATLARI VE FAİZ ORANLARI ARASINDAKİ İLİŞKİNİN ANALİZİ

Today, the size of the mortgage-backed asset market shows us that the housing market can profoundly affect not only the real economy but also the financial sector. The crisis in 2008 is a proof of this circumstance. Compared to developed countries, there is no deepened financial market for mortgage-backed assets in Turkey. However, in Turkey, which implements a growth strategy based on the construction sector, interventions in the housing market have become a part of the economic policy to keep the sector alive. For this reason, policymakers are trying to keep housing loan rates low to differentiate them from other interest rates. However, the effect of these interventions on housing loan rates on housing demand and housing prices is ambiguous. Therefore, the main objective of this study is to investigate the effects of housing loan interest rates on house prices in Turkey in the period of 2010-2020. To analyze the causality between housing loan rates and house prices indexes, we applied Hacker and Hatemi-J (2010) bootstrap Toda-Yamamoto causality test and Hatemi-J (2012) time-varying symmetric Toda-Yamamoto test. The main finding of the study is that there is no causality from the housing loan interest rates to the housing price index in the related period.

An evaluation of competing methods for constructing house price indexes: The case of Warsaw

Accurate house price indexes allow governments and planners to make more effective decisions with regard to land use policy. We compare different ways of computing house price indexes for Warsaw over the period 2006–2018 using a detailed micro-level dataset of 101,182 transactions. We find that when a hedonic approach is used, the resulting index is reasonably robust to the choice of method. More problematic is the repeat-sales method, which is widely used in the US. We find that repeat-sales price indexes are unreliable, affected by sample-selection bias, and prone to significant revisions when new periods are added to the dataset. Even for the hedonic approach, the choice of method can become important for smaller datasets. In such cases, the hedonic time dummy and rolling time dummy methods tend to perform better than the hedonic repricing, hedonic imputation and average characteristics methods. In comparison with the index produced by the National Bank of Poland (NBP), we find that in 2006–7 prices did not rise as much, while since the end of 2012 prices have risen faster than indicated by the NBP index. We attribute these differences to a combination of the choice of hedonic method (we use time-dummy while the NBP index uses the repricing method) and different datasets (the NBP dataset is smaller, especially in the earlier years of the sample).

House price dynamics and relations with the macroeconomic indicators in Turkey

Purpose The purpose of this study is to make an analysis of the short- and long-term effects of inflation, exchange rate, housing interest rate, industrial production index, total housing loans and housing volume on housing inflation in Turkey, taking into account the multiple structural breaks. Design/methodology/approach Multiple structural break Lee–Strazicich unit root test, autoregressive distributed lag bound test and Granger causality test based on error correction model were used. Findings There is both a short- and long-term relationship between housing prices and macrovariables. Housing prices are mostly affected by housing interest rates, housing volume, real exchange rate and total housing loans in the short run. In the long run, it is mostly affected by total housing loans, housing volume and housing interest rates. Research limitations/implications The variables used in the analysis are: housing price index, consumer price index, dollar rate, housing interest rate, industrial production index, total housing loan amount and domestic loan volume. Because the data that variables started common is 2010:M01, the period starting from this date until 2021:M12 is considered. The research covers only Turkey as a country. Determining the micro- and macroeffects of housing prices can always offer solutions for the problems experienced in housing supply and housing demand. Originality/value While investigating housing prices, there are no studies in which total housing loans and housing volume are included in the model together. However, it is important to analyze the effect of the current conjuncture, in which there has been constant increases in foreign exchange rates and high inflation in recent years, on housing prices in Turkey. In this study, investigating these effects by using econometric methods that include structural breaks also increases the original value of this study.

개별 주택가격 데이터를 이용한 주택가격 변동률 분포의 특성 분석*

This study departs from the convention of using only the house price index when analyzing the housing market price change, and examines the market through the lens of house price change distribution. By using a panel data of individual house prices, I find the following. First, house price change distribution has a positive skewness; so, the median is smaller than the mean. Second, when the price index falls, the individual price also falls; however, when the price index rises, the deviation between individual price change increases. Third, even when the price index rises by 10%~20%, 5.2% of the individual houses experience a price fall. Last, the large, high price houses in the capital area shows high price volatility; so, when the price index rises, their price rises more than the average houses, and when the price index falls, their price falls more than the average houses. The results of this study can be utilized in risk management in financial institutions.

Contemporaneous causality among residential housing prices of ten major Chinese cities

PurposeThis study aims to investigate dynamic relationships among residential housing price indices of ten major Chinese cities for the years 2005–2021.Design/methodology/approachUsing monthly data, this study uses vector error correction modeling and the directed acyclic graph for characterization of contemporaneous causality among the ten indices.FindingsThe PC algorithm identifies the causal pattern and the Linear Non-Gaussian Acyclic Model algorithm further determines the causal path, from which this study conducts innovation accounting analysis. Sophisticated price dynamics are found in price adjustment processes following price shocks, which are generally dominated by the top tiers of cities.Originality/valueThis study suggests that policies on residential housing prices in the long run might need to be planned with particular attention paid to these top tiers of cities.

HOUSE PRICE PREDICTION

Because house prices rise every year, a mechanism to forecast future house values is required. House price prediction can help the developer determine the selling price of a house and can help the customer to arrange the right time to purchase a house. Physical conditions, concept, and location are only a few of the aspects that determine the price of a home. Usually, House price index represents the summarized price changes of residential housing. While for a singlefamily house price prediction, it needs a more accurate method based on location, house type, size, build year, local amenities, and some other factors which could affect house demand and supply. A practical and composite data pre-processing, creative feature engineering method is investigated with limited dataset and data features. This model is used to predict the house prices so as to cut down the complications faced by the customers. The present method where customers reach real-estate agents and search for houses in their budget, and should analyze whether a particular price is accurate or not. To overcome this our proposal is used. This system makes optimal use of the Machine Learning Algorithms. By extracting data from datasets of different houses, preprocessing the data and model is built using that data using Regression. The algorithm used for the model building is KNN (K Nearest Neighbor) Algorithm. This system design is modularized into various categories.

Impact of the COVID-19 pandemic on the housing market at the epicenter of the outbreak in China.

The outbreak of the COVID-19 in January 2020 has had a profound impact on the global economy, so it is important to study the impact of the pandemic on the housing market. To investigate the impact of the pandemic on the housing market and the response of the housing market, this paper first uses the hedonic price model to compile the second-hand housing price index in Wuhan and its neighboring capital cities and then uses the difference-in-difference (DID) model to conduct a comprehensive study on new commercial housing and second-hand housing market. In addition, this paper also uses the VAR model to explore the housing market’s response to the epidemic situation. The results show that the negative impact of the pandemic on the housing market is mainly reflected in the volume and area of housing transactions, with little impact on housing prices. Second, the reported cases of COVID-19 have a negative impact on the housing market in the short term, which gradually weakens with time and disappears after three weeks. This paper’s findings indicate that the epidemic’s impact on the housing market is mainly due to the real estate enterprises stopping selling houses and local governments implementing home quarantine measures, which affect normal housing transactions. However, the COVID-19 pandemic did not greatly negatively impact consumers’ demand and confidence in buying houses, so the house prices remained stable overall.Supplementary InformationThe online version contains supplementary material available at 10.1007/s43546-022-00225-2.

Residential housing price index forecasting via neural networks

Residential housing price index forecasting via neural networks

Housing market price movements under tech industry expansion during COVID-19

PurposeThis study aims to analyze the impact of technology-based corporation relocation on housing price indices during COVID-19 within the metropolitan areas of Austin, Texas and Seattle/Bellevue, Washington.The corporations under observation were Tesla and Amazon, respectively. The analysis intends to understand economic drivers behind the housing market and the radius of its effect while including fixed and random effects.Design/methodology/approachThis study used a difference-in-difference (DID) method to evaluate changes in housing price index near and further away from Tesla’s and Amazon’s new corporate locations. The DID method allows for the capture of unique regional characteristics, as it requires a treatment and control group: housing price index and 5-mile and 10-mile search radii centered from the new corporate location.FindingsThe results indicated that corporate relocation announcements had a positive effect on housing price index post-pandemic. Specifically, the effect of Tesla’s relocation in Austin on the housing price index was not concentrated near the relocation site, but beyond the 5- and 10-mile radii. For Seattle/Bellevue, the effect of Amazon’s relocation announcement on housing price index was concentrated near the relocation site as well as beyond a 10-mile radius. Interestingly, these findings suggest housing markets incorporate speculation of prospective economic expansion linked with a corporate relocation.Originality/valuePrevious literature assessed COVID-19 housing market conditions and the economic effects of corporate relocation separately, whereas this study analyzed the housing price effects of corporate relocation during COVID-19. The DID method includes spatial and temporal analyses that allow for the impact of housing price to be observed across specified radii rather than a city-wide impact analysis.

주택가격요인 분석에 대한 연구

The volatility of the housing price based on demand and supply in the housing market reflects the structural features of the housing market. Consequently, decomposing the housing market into trends and cycles through econometric models will make it possible to deduce the nature of the housing market in the long term, in addition to provide hindsight for real estate portfolio asset allocation in a more systematic way.BRBR The paper addresses the housing price behavior through univariate time series analysis. The Unobserved-Component (UC) model and the Beveridge-Nelson (BN) decomposition model are applied to analyze the factors for the volatility of housing price with the trends and cycles. Specifically, the UC model is analyzed by UC-Zero(0) model assuming the covariance between the trends and the cycles, which are unobserved variables, is zero (0), and the UC-UR (UC-UnRestricted) model, which is free from the restraint of the covariance between the trends and cycles. The estimation is based on the Kalman Filter. The study analyzes how the volatility in the domestic housing price index was influenced by the trends and cycles, which are inherent variables, by comparing the estimated values of the UC-0 and UC-UR model with the BN decomposition model. Domestic housing prices have plunged two times: the 1997 Asian financial crisis and the 2008 Global financial crisis. However, the analysis data in this research include only the 2008 Global financial crisis.BRBR The results are summarized as follows: First, the volatility of the domestic housing price index can be explained by the UC-0 model. As explained by the UC-0 model with the assumption of the covariance between trend and cycles zero(0), the domestic housing price index has not shown any correlation between the trend and cycle, and thus it implies that the trends mainly explain the volatility in the housing price index. Second, as the trends mainly explain the volatility in housing prices, domestic housing prices do not follow a random walk under the efficient market hypothesis. Hence, the price volatility can be seen as predictable, and the right-upward price trend that has been lasted for a considerable period of time can be seen as a result of reflecting adaptive expectations of market participants. Lastly, institutional investors such as pension funds are increasing their share of real estate and alternative investments, and the share of real estate is increasing in the asset market. The fact that inefficient housing market, and the housing price return as seen through the housing price analysis follow a very differential distribution from the normal distribution would mean that as a factor hindering investors