研究以房價指數、匯率及貸款利率為研究標的，就台灣與日本兩地，2008年1月至2014年12月，共84個月期間，利用非線性門檻誤差修正模型架構，分別研究台灣、日本的房地產市場與匯率、利率其相互間之長短期非線性因果關係。在研究方法上，採用Kapetanios et al.(2003) KSS非線性單根檢定以測試非線性的定態關係，並以Enders and Granger(1998)門檻自我迴歸模型(TAR)以及動差門檻自我迴歸模型(M-TAR)進行門檻共整合檢定，再進一步利用Enders and Granger(1998)及Enders and Siklos(2001)門檻誤差修正模型(TECM)來捕捉台灣、日本的房地產市場與匯、利率市場之長短期非線性不對稱效果。 從實證結果發現在線性PP、KPSS與NP單根檢定法，檢測中、美股匯資料皆為I(1)數列。而在門檻共整合檢定部份，綜合發現無論是在台灣或日本，房地產與利率皆存在有長期均衡的非對稱共整合關係。最後，由門檻誤差修正模型因果關係檢定綜合發現，在短期，不論台灣或日本的房價與匯率之間的關係皆屬於傳統理論。在長期，台灣的房價與匯率之間的關係屬於傳統理論與投資組合理論，而日本的房價與匯率之間的關係不屬於傳統理論。 This paper empirically investigates the exchange rate effects of the NT dollar against the JPY dollar on residential index in Taiwan and Japan. Second, This paper also investigates the interest rate effects on residential index in Taiwan and Japan This study employs the newly threshold error-correction model (TECM) elaborated by Enders and Granger and Enders and Siklos, assuming the nature of the relationship between the variables is on the basis of non-linearity. The empirical evidence suggests that there is a long-run equilibrium relationship between NTD/US 、JPY/US and the residential index of Taiwan and Japan during the time period investigated. In addition, the results of TECM Granger-Causality tests show a short-run causal relationship exists between NTD/US、JPY/US and the residential index of Taiwan and Japan. However, in the long-run a bidirectional causal relationships between NTD/US and the residential index of Taiwan strongly argues for the traditional approach and portfolio approach. And in the Japan, a bidirectional causal relationships between JPY/US and the residential index of Japan during doesn’t argue for the traditional approach. But there is positive causal relationship from the residential index of Taiwan and Japan to interest rate, and also strongly argues for the portfolio approach.
