본문 바로가기
KERD
논문KCI 등재

A spatio-temporal model of house prices in Korea: a Bayesian approach

박헌수|전해정
한국주거환경학회 (Residential Environment Institute Of Korea)|2018.06|주거환경 (Journal of The Residential Environment Institute of Korea)|제16권 제2호|pp. 73-90|DOI http://dx.doi.org/10.22313/reik.2018.16.2.73

1

조회

0

다운로드

0

인용

0

참고문헌

키워드

GIS지리정보시스템주택가격예측house pricesprediction베이지안 이분산성 시공간자기회귀모형
주제분류국제/지역개발

초록

The purpose of this article is to show how spatiotemporal correlations can be incorporated when estimating regression coefficients and when predicting house prices. The empirical analysis is conducted using standard OLS model and spatial temporal autoregressive models. The data set used in the estimation of the hedonic models consists of transaction data of apartments for Gangnam-gu, Seoul from January 2006 to December 2015. Data processing is facilitated by the use of a vector-based Geographic Information Systems (GIS). By considering spatiotemporal effects, this paper extends Pace et al. (1998)’s spatiotemporal autoregressive model (STAR) with the consideration of the heteroscedasticity problem arising from the nature of the data. The family of the spatio-temporal autoregressive models produces better goodness of fit and out-of-sample performance than the family of the traditional hedonic models based on OLS with and without trend models. Our Bayesian heteroscedastic STAR model has significantly increased the R^2 to 0.8725, increased by 230.2% relative to OLS without trend model and increased by 103.7% relative to OLS with trend model. It also has significantly reduced root mean square errors (RMSE) and the median absolute errors. Based on the results of this study, if the house price is estimated by using the BSTAR model, the government policy makers can accurately understand the movements of the housing market and can establish and implement policies and investors will be able to invest objectively.

Abstract

The purpose of this article is to show how spatiotemporal correlations can be incorporated when estimating regression coefficients and when predicting house prices. The empirical analysis is conducted using standard OLS model and spatial temporal autoregressive models. The data set used in the estimation of the hedonic models consists of transaction data of apartments for Gangnam-gu, Seoul from January 2006 to December 2015. Data processing is facilitated by the use of a vector-based Geographic Information Systems (GIS). By considering spatiotemporal effects, this paper extends Pace et al. (1998)’s spatiotemporal autoregressive model (STAR) with the consideration of the heteroscedasticity problem arising from the nature of the data. The family of the spatio-temporal autoregressive models produces better goodness of fit and out-of-sample performance than the family of the traditional hedonic models based on OLS with and without trend models. Our Bayesian heteroscedastic STAR model has significantly increased the R^2 to 0.8725, increased by 230.2% relative to OLS without trend model and increased by 103.7% relative to OLS with trend model. It also has significantly reduced root mean square errors (RMSE) and the median absolute errors. Based on the results of this study, if the house price is estimated by using the BSTAR model, the government policy makers can accurately understand the movements of the housing market and can establish and implement policies and investors will be able to invest objectively.

참고문헌 (30)

  1. Anselin, L., "Spatial Econometrics:Methods and Models" (1988), Kluwer Academic
  2. Basu, S., "Analysis of Spatial Autocorrelation in House Prices", Journal of Real Estate Finance and Economics, 17(1), pp.61~85 (1998)
  3. Beamonte, A., "Bayesian inference in STAR models using neighbourhood effects", Statistical Modeling, 8(3), pp.285~311 (2008)
  4. Beamonte, A., "Robust Bayesian inference in STAR models with neighbourhood effects", Journal of Statistical Planning and Inference, 140, pp.3047~3057 (2010)
  5. Can, A., "Specification and Estimation of Hedonic Housing Price Models", Regional Science and Urban Economics, 22, pp.453~474 (1992)
  6. Can, A., "The Measurement of Neighborhood Dynamics in Urban Housing Price", Economic Geography, 66(3), pp.254~272 (1990)
  7. Can, A., "Spatial Dependence and House Price Index Construction", The Journal of Real Estate Finance and Economics, 14, pp.203~222 (1997)
  8. 전해정, "시공간자기회귀(STAR)모형을 이용한 부동산 가격 추정에 관한 연구", 부동산연구, 24(1), pp.7~14 (2014), 한국부동산연구원
  9. Dubin, R. A., "Predicting House Prices Using Multiple Listings Data", The Journal of Real Estate Finance and Economics, 17(1), pp.35~59 (1998)
  10. Dubin, R. A., "Estimation of Regression Coefficients in the Presence of Spatial Autocorrelated Error Terms", Review of Economics and Statistics, 70, pp.466~474 (1988)
  11. Farber, S., "Topology and dependency tests in spatial and network autoregressive models", Geographical analysis, 41(2), pp.158~180 (2009)
  12. Gelfand, A. E., "The Dynamics of Location in Home Price", The Journal of Real Estate Finance and Economics, 29(2), pp.149~166 (2004)
  13. Gelfand, A. E., "Spatio-Temporal Modeling of Residential Sales Data", Journal of Business and Economic Statistics, 16, pp.312~321 (1998)
  14. Geweke, J., "Bayesian Treatment of the Independent Student Linear Model", Journal of Applied Econometrics, 8, pp.19~40 (1993)
  15. 김성우, "공간계량모형에서의 실제거리를 반영한 공간가중행렬에 관한 연구 -부산아파트 실거래가를 중심으로-", 주택연구, 18(4), pp.59~80 (2010), 한국주택학회
  16. Kim, C. W., "Measuring the benefits of air quality improvement: a spatial hedonic approach", Journal of Environmental Economics and Management, 45(1), pp.24~39 (2003)
  17. 이창로, "토지 가격의 베이지안 시공간 분석", 부동산연구, 24(4), pp.119~135 (2014), 한국부동산연구원
  18. LeSage, J. P., "Unpublished Manuscript" (1999), University of Toledo
  19. McMillen, Daniel P., "One Hundred Fifty Years of Land Values in Chicago: A Nonparametric Approach", Journal of Urban Economics, 40, pp.100~124 (1996)
  20. Militino, A., "Alternative models for describing spatial dependence among dwelling selling prices", Journal of Real Estate Finance and Economics, 29(2), pp.193~209 (2004)
  21. Mizruchi, M. S., "The effect of density on the level of bias in the network autocorrelation model", Social Networks, 30(3), pp.190~200 (2008)
  22. Nappi-Choulet, Ingrid Pr., "A Spatiotemporal Auto regressive Price Index for the Paris Office Property Market", Real Estate Economics, 27(2), pp.305~340 (2009)
  23. Pace, R. K., "A method for spatial temporal forecasting with an application to real estate prices", International Journal of Forecasting, 16, pp.229~246 (2000)
  24. Pace, R. K., "Spatiotemporal Autoregressive Models of Neighborhood Effect", The Journal of Real Estate Finance and Economics, 17(1), pp.15~33 (1998)
  25. Pace, R. K., "Generalizing OLS and the Grid Estimator", Real Estate Economics, 26(2), pp.331~347 (1998)
  26. Rosen, S., "Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition", The Journal of Political Economy, 82(1), pp.34~55 (1974)
  27. Smith, T. E., "Estimation bias in spatial models with strongly connected weight matrices", Geographical analysis, 41(3), pp.307~332 (2009)
  28. Stein ML, "Predicting random fields with increasingly dense observations", Annals of Applied Probability, 9, pp.242~273 (1999)
  29. Takatsuka, H., "A Present-Value Model of Real Estate with Interneighborhood Dependency of Incomes", Journal of Real Estate Finance and Economics, 23(1), pp.47~76 (2001)
  30. Wilhelmsson, M., "Spatial models in real estate economics", Housing, Theory and Society, 19(2), pp.92~101 (2002)

이 간행물의 다른 논문 (29건)

都市臨海部の共同住宅における水辺空間に対する意識に関する研究야마치카, 김주홍pp. 1-10공인중개사에 대한 사회적 인식 제고에 관한 연구윤동건, 오세준pp. 11-24서울시 오피스 빌딩 매매가격 결정요인에 관한 연구-금융위기 이전과 이후를 구분하여-윤영식, 이새나리pp. 25-42경제자유구역의 효율적 추진방향에 관한 연구－ 대구 수성의료지구를 대상으로 －김철영, 서준교pp. 43-55대지진에 의한 건축물의 재해 패턴 분석신상화, 김진, 박강근pp. 57-72
원문 보기