ARIMA 모형을 이용한 전국 아파트시장 가격 예측

Abstract

The purpose of this study is to make a model to forecasting apartment trade price and household rent price in apartment housing market using ARIMA model. On the basis of those models, I tried to forecast the fluctuations of short-term apartment trade price and household rent price in national apartment market.
 To analyze the ARIMA model, quarterly data during 2011 1/4∼2023 3/4 are used for identification, estimation, diagnosis, and prediction of the ARIMA model. Using ARIMA model, the outcome ARIMA(2,1,1) model is applied to nationwide apartment trade price in the rate of apartment trade price forecasting model, and ARIMA(1,1,1) model is applied to nationwide apartment household rent price in the rate of apartment rent price.
 According to the forecast results of nationwide apartment trade price and apartment household rent price in the ARIMA model, looking at the quarterly rate of change in national apartment sales prices, it appears that they will continue to fall in the 2023 4/4, but will turn to an upward trend starting in the 2024 3/4, and will continue to rise slightly but without significant fluctuations. Looking at the quarterly trend of national apartment rental prices, the downward trend is expected to turn into an upward trend in the 2023 4/4 and continue until the 2024 4/4, but the upward trend will not be large.
 While the government's housing policy is sluggish, national apartment sales prices and Jeonse prices are expected to be affected by a decrease in apartment supply and interest rates due to a decrease in permits and occupancy next year.