Apartment Transactions Research Articles

Towards a Better Uncertainty Quantification in Automated Valuation Models

AbstractThis study introduces a novel framework for quantifying prediction uncertainty in automated valuation models (AVMs), crucial tools in modern real estate finance. While non-linear AVMs excel in predictive performance, their limited methods for assessing prediction uncertainty reduces reliability and practical utility. We address this gap by proposing an approach for quantifying the uncertainty associated with predicted house prices and by introducing a model-specific AVM uncertainty estimate (AVMU) for AVM comparisons. Using a dataset of 51,747 historical apartment transactions in Oslo, Norway, we train three AVMs (XGBoost, random forest, support vector machine) to predict sales prices. Thereafter, we develop three base uncertainty estimators (direct loss estimation, bootstrap ensemble, quantile regression) and three meta estimators (average regressor, voting regressor, stacked generalization) for uncertainty quantification. Conformal calibration aligns the outputted uncertainty estimates from the six estimators with standard deviations of corresponding prediction errors. Having strong positive correlations with observed absolute prediction errors, the calibrated uncertainty estimators are shown to effectively capture prediction uncertainty. While the direct loss estimation excels among base estimators, the voting regressor and stacked generalization meta estimators consistently outperform it. Furthermore, by using the AVMU estimate from the stacked generalization meta estimator we can successfully identify the best-performing AVM for three separate apartment portfolios without knowing true sales prices. This alignment of the mean estimated prediction uncertainty with observed deviations underlines the utility of pre-factual AVMU estimates for model comparisons. In conclusion, our framework helps bridge prediction accuracy and uncertainty for AVMs, enhancing their reliability and supporting informed decision making for stakeholders.

법인 거래량 비율이 아파트 가격에 미치는 영향

This study focused on the steadily increasing proportion of corporate transactions in the total apartment transaction volume. It also investigated the correlation between this proportion and apartment prices. The analysis confirmed a positive relationship between the corporation's share in apartment transaction volume and the apartment sales index. In particular, the direction of this relationship varied depending on the corporation's market participation status. When a corporation participated as a seller (buyer), it positively (negatively) correlated with the apartment sales index. This study has implications for achieving the policy goal of curbing speculative demand from corporations. However, the system must be restructured to incentivize corporations to actively engage in real estate transactions as sellers to ensure price stability in the real estate market.

수도권광역급행열차(GTX)-B노선 추진에 따른 사업단계별 주변 아파트 가격에 미치는 영향에 관한 연구

This study is a study on the effect of the GTX B route departure station on the price of nearby real estate around Songdo International City, Incheon University Entrance Station, as the GTX B route was confirmed, and 10,142 transaction cases were collected from July 2016 to September 2020 to analyze the effect on the prices of nearby apartments by stage of the GTX B route project according to the preliminary feasibility study and selection through the difference-in-differences method. As a result of the analysis, it was confirmed that the second station influence area was affected before and after the preliminary feasibility study was selected, and it has a significant impact before and after the preliminary feasibility test, but this is not a policy effect, and when the individual characteristics of the apartment are included, the coefficient difference is large, so the robustness is poor. In the preliminary feasibility test, the unit price of apartment transactions is considered to be largely affected by the housing characteristics effect of the apartment and the macroeconomic effect that the entire society is affected. Therefore, this paper analyzes the price changes of surrounding apartments by project stage following the GTX B route promotion, presents future changes in apartment prices near the GTX B route pre-jeong station, and further research should be conducted in accordance with the analysis of various factors such as interest rates and the government’s real estate policy in the future and the promotion of additional projects such as the construction and opening of the GTX B route.

Multiscale Impacts of Land Environmental Features and Planning on Apartment Resale Prices in Jinan City, China

As a typical city with a population of 5 to 10 million in China, Jinan has undergone significant increases in land supply during the past decade, resulting in substantial volatility in apartment sale/resale market prices. In this study, we investigated all second-hand apartment transactions from 826 communities of Jinan city and explored the multiscale impacts of land environmental features and planning on apartment resale prices throughout the city. Specifically, central and eastern regions had significantly positive impacts on apartment resale prices, while western regions had significantly negative impacts; education resources had consistently positive impacts throughout the city while shopping, business buildings, and medical resources had insignificant impacts; subway stations had insignificant impacts and bus stations had significant effects only in congestion points and northeastern edges. Our results revealed the formation mechanisms and spatial heterogeneity of apartment resale prices in Jinan. Our work will not only help in the decision making of potential apartment purchasers, but will also be conducive to enhancing the spatial justice of local governments in land supply and planning policies.

Measuring spatial heterogeneity of air quality on apartment transaction prices in Seoul, South Korea

Measuring spatial heterogeneity of air quality on apartment transaction prices in Seoul, South Korea

Still prefer, but not that much: the premium of subway access on house prices during the COVID-19 pandemic period

ABSTRACT Rail access has traditionally been considered a factor that enhances the amenity value of urban areas, resulting in a price premium for properties located near these stations. However, the Covid-19 pandemic has exerted a profound influence on people's mobility patterns and travel behavior within urban areas. One of the most conspicuous changes is a substantial decrease in subway ridership. This study investigates whether the decline in public transit ridership has led to shifts in housing location preferences driven by proximity to rail access during the Covid-19 pandemic. This study focuses on assessing the impact of the pandemic on the premium associated with rail access in relation to house prices. Using approximately 180,000 apartment transaction records from Seoul, South Korea, the empirical findings, based on a difference-in-differences approach, reveal an immediate and significant depreciation effect resulting from this exogenous shock during the 25-month pandemic period. The premium decreases significantly within a 1.2-kilometer radius from a subway station, and this reduction exhibits a distance-decay pattern within the designated area. Highlights Traditional amenity value of rail access challenged by Covid-19. Profound decline in subway ridership during pandemic. Investigation into pandemic's impact on housing location preferences. Assessment of pandemic's effect on rail access premium. Immediate and significant depreciation effect within 1.2 km from subway stations.

지방 아파트 가격 결정 요인과 특성 분석

1. CONTENTS
 (1) RESEARCH OBJECTIVES
 The purpose of this study is to identify the characteristics that influence the determinants of apartment prices between high and low apartment prices within the same autonomous district in the regional submarket, and to further confirm whether there is a price gap between branded and non-branded apartments.
 (2) RESEARCH METHOD
 For this purpose, a multiple regression analysis was conducted targeting all apartments traded over the past three years in Buk-gu, Busan (Hwamyeong-dong, Mandeok-dong), using the actual apartment transaction price as the dependent variable and independent variables related to apartment structure, complex, location, and brand characteristics.
 (3) RESEARCH FINDINGS
 As a result of the analysis, the variables that commonly had a significant positive (+) influence on the actual transaction price were the year of occupancy and the number of apartments, while the variables that had a negative (-) influence were the exclusive use area. In Hwamyeong-dong, a high-price area, the distance to elementary schools, convenience facilities, and apartment brands had a positive (+) effect, while in Mandeok-dong, a low-price area, they had a negative (-) effect.
 2. RESULTS
 This means that there may be a gap in apartment prices even within the same area, and even for non-branded apartments, the price may be higher than branded apartments if the area is close to a station or close to convenience facilities. In this study, these results were presented as marketing Unique Selling Point related to the supply of regional and non-branded apartments.

Economic effects of air quality on housing prices: evidence from Beijing, China

Air pollution is a major environmental urban issue, particularly in fast-growing cities in developing countries. Reducing air pollution is thus a challenge while evaluating the economic value of air quality is crucial for environmental policies made. However, few studies accurately estimate this value as they neglect the possible endogeneity issues, as well as the dynamic and heterogeneous effects of air pollution. Under the hedonic framework, we therefore assess the economic effect of fine particulate matter (PM2.5) on housing prices in Beijing, China. We construct a panel based on resale apartment transactions matched with average quarterly PM2.5 data between 2013 and 2019. To reduce the risk of an estimation bias, we apply an instrumental variable (IV) approach. Our results show that PM2.5 is negatively associated with housing prices. Households were willing to pay an extra 0.0852% per housing unit price for an average quarterly reduction in PM2.5 of 1 µg/m3. Furthermore, we argue that high-income dwellers tend to pay more for clean air. The negative effects of PM2.5 across regions are significant and different. Compared with that in the basic year 2013, the negative effect increases in the first 3 years and then decreases in the last 3 years. Our findings enhance our comprehension of the economic impact of air quality and make a valuable contribution to the nuanced understanding of willingness to pay for air quality, which is beneficial in assessing and optimizing environmental regulations.

Does the renovation of waste-to-energy incineration plants attenuate the housing price gradient? Evidence from a quasi-natural experiment in Shenzhen, China

Waste-to-energy (WTE) incineration plants, which inevitably cause negative externalities due to their impacts on the surrounding environment, have encountered frequent protests from residents during their construction and operation. The negative externalities could lead to property value losses by impacting residents' marginal willingness to pay (MWTP). To assess whether the renovation of WTE incineration plants can attenuate the housing price gradient around the plants, we use the completion of the renovation project of Yantian Plant in Shenzhen as a quasi-natural experiment to exploit difference-in-differences (DID) estimators, and propose a hedonic price model based on resale apartment transaction data within 10 km of the plant. Results indicate that the renovation completion narrows the price gap between proximate and distant apartments from the plant by 35.25%, and the housing price gradient begins to decrease significantly in the late construction period. Moreover, the effect on the housing price gradient remains sustained throughout the whole post-treatment period. Furthermore, the analysis of the underlying mechanism suggests that the renovation completion attenuates the housing price gradient through the improved environment.

AI 스마트 하우징 구축이 서울시 아파트 거래가격에 미치는 영향

The AI smart housing market is expected to grow, but there is no study on the effect of the installation of AI smart housing on housing prices in terms of activation. It was attempted to demonstrate whether there is a housing price as a characteristic of a house through the hedonic model, which is most commonly used.
 The empirical data was used by combining the K-apt apartment management fee disclosure data and the apartment price and apartment characteristics data provided by Real Estate 114 Reps. According to the apartment management cost information provided by K-apt, 2,144 (12.15%) of 17,638 apartment complexes nationwide for which the AI Smart Housing maintenance fee is charged as an AI Smart Housing were installed. In the case of Seoul, an AI Smart Housing was installed in 213 (8.59%) of the total 2,479 apartment complexes.
 As a result of attaching price information to the management cost information, a total of 12,073 apartment transaction price data were used, and 1,598 pieces of AI Smart Housing installation complexes were used for the analysis. As a result of the empirical analysis, as in the general hedonic model, the house sale price per unit area was higher as the transaction level increased, the number of years elapsed, the total number of households, and the common management cost per household increased. In the case of common management fees, in the case of apartments that are recently supplied, the common management fees are high according to the trend of supplying various resident convenience facilities such as swimming pools and golf driving ranges in the complex, as well as large lobby spaces, and this can be seen as a result of this part being reflected in the house sale price. . In addition, the fact that the smaller the exclusive area is, the higher the price per unit area is due to the fact that the preference for small flats due to the downsizing of the household size, such as the increase in one- or two-person households, is reflected. Lastly, it was found that the apartment price per unit area was higher when the AI Smart Housing, the subject of this study's interest, was installed, compared to the case where the AI Smart Housing was not installed.

Adverse selection in iBuyer business models—don’t buy lemons!

The rise of instant buyer (iBuyer) businesses in the past years has made automated valuation models (AVMs) an important part of the property market. Although iBuyer services are in demand, large actors within the segment have reported dissatisfying profits over time. The business model is subject to adverse selection as homeowners based on their superior knowledge of their home are more likely to accept overpriced bids than underpriced bids, making the iBuyer purchase more overpriced dwellings. In this paper, we use a dataset consisting of 84,905 apartment transactions from Oslo, the Norwegian capital. We use 80% of the dataset to train three different AVMs similar to those used by iBuyers. Next, we construct some simple purchasing rules from the predictive accuracies found in the training dataset. Finally, taking the remaining 20% of the data in a test dataset, we introduce an adverse selection indicator based on accepted probability distributions and calculate the average expected resale profits per apartment for a hypothetical iBuyer. We find that adverse selection has a large negative impact on average profits for the hypothetical iBuyer. Furthermore, the simple purchasing rules are able to improve the profit by 1 percentage point per apartment when adverse selection is present.

COMBINATION OF MACHINE LEARNING-BASED AUTOMATIC VALUATION MODELS FOR RESIDENTIAL PROPERTIES IN SOUTH KOREA

The applicability of machine learning (ML) techniques has recently been expanding to include automatic real estate valuation models. The main advantage of this technique is that it can better capture complexity in the value determination process. Therefore, the performance of these techniques is shown to be superior to conventional models. In this paper, the latest ML algorithms (i.e., support vector machine, random forest, XGBoost, LightGBM, and CatBoost algorithms) are examined as automatic valuation models, and several combination methods are proposed to improve the models’ predictive power. We applied ML models to approximately 57,000 records on apartment transactions, which were provided by South Korea’s Ministry of Land, Infrastructure, and Transport, that occurred in Seoul in 2018. The results are as follows. First, ML-based predictors (especially, the latest decision tree-based algorithms) are more performative than conventional models. Second, the prediction error from a model can be partially offset by another model’s error, which implies that an efficient averaging of the predictors improves their predictive accuracy. Third, the models’ relative performance may be relearned by the ML algorithms, which means that they can also be used to recommend which algorithm should be selected for making predictions.

출산율과 인구구조의 변화가 주택가격에 미치는 효과: 동탄신도시를 중심으로

As a result of analyzing the influence of variables related to the demographic structure, it was found that the price of apartments traded at a time when the total fertility rate was high, the lower the price. In the case of acetic age, the price of apartments traded at a time when the acetic age was higher was found to be lower. It was found that the ratio of single-person households had a negative effect on apartment transaction prices, and the ratio of double-person households had a positive effect on apartment transaction prices. These results suggest that transaction prices tend to fall as apartments traded at a time when the number of single-person households composed of unmarried or students increases. On the contrary, it was found that apartments traded at a time when more double-person households move in are more likely to rise in price. The elderly ratio was found to have no significant effect on apartment transaction prices, but the higher the price of apartments traded at a time when more people in their 60s moved in. In addition, it was found that the high proportion of people in their 60s moved in and apartment prices rose during the period when their transactions were concentrated, mainly in high-priced apartment complexes. These results suggest that changes in fertility rates and demographic structure have a significant impact on apartment prices in new cities. Therefore, it seems important to predict changes in the fertility rate and demographic structure of the region when deciding on apartment supply policies to prevent soaring apartment prices.

평택시의 권역별 개발사업이 주택가격에 미치는 영향에 관한 연구

(1) RESEARCH OBJECTIVES The purpose of this study is to study how the real estate development project of Pyeongtaek City affects housing prices by subdividing the northern, southern, and western regions by region. (2) RESEARCH METHOD This study employed the Hedonic Price Model to achieve the major objectives of this study. (3) RESEARCH FINDINGS The implications of this study are as follows. First, through this study, it was found that although the start of development projects raised the housing price, it did not falls within the spatial scope of this study, was judged to be meaningful to examine the differences by region as a place where the differences by region are clearly visible. 2. RESULTS The results of this study are as follows. The influence of other characteristics did not differ significantly by region, but in terms of development characteristics, it was found that the SRT Jije Station start and Samsung Electronics start construction increase the apartment transaction price in the southern and western regions except for the northern region.

Effects of the Olympic Announcement and the Actual Event on Property Values: The Case of the 2018 PyeongChang Winter Olympic Games

This study investigates the effects of the Olympic announcement and the actual event on property values in the host region using the case of the 2018 PyeongChang Winter Olympic Games. We collected Korean government-issued apartment transaction data in the host city, Gangneung, and other cities. We performed propensity score matching to generate a suitable control group compared to units in Gangneung and exploited difference-in-difference analyses to test the impact of the announcement and the actual event separately. The results show that the Olympic announcement increased property values in Gangneung by 5.5% compared to propensity-matched units, and the effects are mostly shown in units in downtown Gangneung. Units close to the KTX station and the Olympic arena observed additional increases in housing prices. During the actual Olympic event, property values in Gangneung increased by 8.3% after the KTX station opened, but the actual event period and Athletes’ Village opening did not generate additional effects. The additional increases in housing prices close to the opening of the KTX station were found for units close to the KTX station, the Olympic arena, and Athletes’ Village.

Assessing the economic premium of electric vehicle charging points: a hedonic price analysis in Beijing, China

Electric vehicle charging points (EVCPs) are the foundation of the EV industry and are conducive to low-carbon transportation development. Research on EVCPs mainly focus on operation profits and layout, while no study assesses their economic premium. Aiming at providing a reference for the installation of EVCPs, this study employs a hedonic price model to evaluate the economic premium of CPs, based on apartment transactions and complex data on CPs in 2015 and 2016 in Beijing. The primary findings indicate the economic premium of CPs is significantly positive in both housing sale and rental markets. However, the impact of CPs on housing prices is heterogeneous in terms of complex floor area ratio and location. Specifically, the premium effect of CPs is more significant in larger complex and complex in the suburb. Following the above conclusions, we provide several policy pathways to promote the construction of CPs and their scientific location layout.

아파트 대량 평가를 위한 기계 학습 알고리즘과 스태킹 앙상블 기법의 응용

Purpose The goal is to find an optimal data-driven model for apartment price appraisal system. We compares the predictive power of several machine learning algorithms and the ensemble model by using 17,779 apartment transaction data in Gangnam from 2006 to 2017. Methods We use three tree-based machine learning techniques, random forest, Adaboost, and Catboost. Also, an ensemble model is developed through the stacking technique using the above algorithm. 5-fold cross validation is used as model validation techniques. Results By applying the algorithm, it was found that the ensemble model constructed from stacking method performs better than individual machine learning models. Conclusion The results illustrate that the quality of mass appraisals or apartment price indices can be improved by using the ensemble model.

Predicting the Apartment Transaction Price by Using Machine Learning Methods and the Fixed Effects Model

Predicting the prices of housing is a critical issue for all real estate market participants, including financial institutions and governments, as well as homeowners and potential buyers. To evaluate the performance of a housing price prediction analysis model, this study compares machine learning methods and the fixed effects model based on the panel data of actual transactions between January 2006 and July 2020 in Seoul. For this purpose, we are also considering the characteristics of the apartment complex, based on the hedonic pricing approach, as well as macroeconomic variables and policy variables. We find that the Random Forest (RF) model shows higher predictive accuracy than the Multivariate Adaptive Regression Splines (MARS) model in machine learning. Our study shows the potential of the wide use of the machine learning method by utilizing micro-data in predicting and forecasting housing prices more accurately.

The Effect of a New Subway Opening on House Prices Near Metropolitan Bus Stops in a New Town: The Case of the New Town of Gwanggyo

This study examines the effect of a new subway opening on house prices near metropolitan bus stops, which provide the only means of transportation to Seoul before reaching the subway, in a new town. In the empirical analysis, a difference-in-differences is applied on the Gwanggyo housing market, where the Shinbundang Line has opened in 2016. Using over 4,800 apartment transaction data, empirical results suggest that the opening of the Shinbundang Line has a positive impact on house prices near metropolitan bus stops. Apartment prices increase within 300 m from metropolitan bus stops by up to approximately 10.9% after the opening of the Shinbundang Line. The results imply an overall increase in housing demand in the region due to the expansion of the transportation network. Likewise, the effects on price tend to be more significant in the long term, like in three years’ time, than in the short term, like in one and in two years’ time.

A Study on Developing the Regional Apartment Transaction Concentration Index

A Study on Developing the Regional Apartment Transaction Concentration Index