서울 아파트 전세시장의 순환주기 분석: 고속 푸리에 변환과 연속 웨이블릿 변환을 이용한 시계열 접근

This study conducted a multi-layered time series analysis applying fast Fourier transform (FFT) and continuous wavelet transform (CWT) to investigate the medium and long-term cycle structure inherent in the monthly jeonse price index of apartments in Seoul. After removing the long-term trend, the average cycle pattern was confirmed through FFT analysis, and the 42-month (3.5 years) cycle was the strongest cycle structure. In addition, multiple cycles of around 36, 48, and 60 months were identified as meaningful cycle components. As a result of the CWT analysis, the 40-60-month cycle was repeatedly observed throughout the analysis period, and the mid- to long-term cycle was strengthened in conjunction with changes in the policy and financial environment after July 2020. These results suggest that the Seoul jeonse market is a complex cycle market with a multi-layered medium- and long-term structure rather than a single cycle, and that the volatility and persistence of the jeonse market have been readjusted to the mid- to long-term cycle level after the policy shock, and suggest that when monitoring the jeonse market and responding to policies, not only short-term indicators but also the actual moving cycle and the mid- to long-term cycle structure linked to institutional changes should be considered together.