주간단위 서울시 아파트 실거래가격지수개발에 관한 연구

초록

This study constructs a weekly transaction-based apartment price index for Seoul using publicly available real estate transaction data from the Ministry of Land, Infrastructure and Transport (MOLIT). A Bayesian repeat-sales model, which has previously been applied to thinly traded markets such as office and land markets, is employed to construct the index at a weekly frequency. To identify an appropriate methodology for constructing a weekly price index, alternative model specifications and assumptions are systematically compared. Based on these comparisons, an autoregressive (AR) process is selected to capture the temporal dependence of housing prices, whereby current prices are related to their past values. The resulting index provides stable estimates and exhibits price movements broadly consistent with those of existing asking price indices. The weekly transaction-based price index developed in this study can complement existing weekly asking price indices by providing transaction-based information on housing price movements and facilitating the ex-post assessment of housing market conditions.

키워드