| 摘要: | 本研究以英國宣告脫歐之英鎊，與特別提款權各通貨間為研究標的，以研究時間為2016年1月1日至2017年2月28日，以英國宣告脫歐日2016年6月24日前後作探討，利用非線性共整合分析英國宣告脫歐前與宣告後之英鎊，與特別提款權各通貨間之長期均衡關係，且利用Granger因果關係及非線性門檻誤差修正模型，分別探討英國宣告脫歐前與脫歐後之英鎊，與特別提款權之美元、歐元、人民幣及日元間的長短期非線性效果及「領先-落後」關係。
首先，利用線性單根檢定(ADF、PP、KPSS)及非線性單根檢定(KSS)檢驗時間序列資料是否穩定，在資料結果不一致情況下，為了讓資料結果強穩本文以KSS檢定得知變數資料皆為I(1)數列。接著進行門檻共整合檢定，不論英國宣告脫歐前或宣告後，美元指數、歐元、人民幣及日元對英鎊互動具有長期非線性共整合關係。最後，變數間互動存在長期非線性門檻共整合關係下，進行門檻修正誤差模型。不論在英國宣告脫歐前或宣告脫歐後，短期互動下，除了宣告脫歐後之傳統ECM模型中，英鎊對人民幣存在單向因果關係外，其餘變數在傳統ECM模型或門檻誤差修正模型中互動皆不存在因果關係；在英國宣告脫歐前，長期互動下，以門檻誤差修正模型，發現美元指數與人民幣對英鎊間走勢偏離程度較小而低於門檻值時，兩變數間變動對英鎊走勢具有長期因果關係。歐元與日元對英鎊間走勢互動在偏離程度大而超過門檻值，兩變數對英鎊匯率趨勢長期下具有單向「領先-落後」關係。在傳統ECM模型得知，英鎊對歐元、人民幣及日元趨勢互動並不存在因果關係；在英國宣告脫歐後，以傳統ECM模型檢驗，得知英鎊對美元指數和人民幣間走勢不具有因果關係。以門檻誤差修正模型檢驗，首先得知，英鎊匯率變動對美元指數不存在長期因果關係。接著，在偏離程度過小而低於門檻值時，英鎊震盪下間接影響歐元走勢。在走勢偏離程度過大而超過門檻值時，人民幣匯率起伏長期間接影響英鎊匯率，可從實證得知人民幣匯率對英鎊走勢具有正向「領先-落後」關係。最後，從日元匯率與英鎊間走勢，兩變數間走勢在偏離程度過大而超過門檻值時，日元與英鎊匯率間走勢在長期互動下具有雙向「領先-落後」關係。
This paper empirically investigates whether asymmetric causal relationship exists between the Pound and Each Currency in the SDR Basket on declaration of Brexit. Moreover, discussing US Dollar Index,Euro, Yuan, JPY and GBP, which use the nonlinear co-integration test to analyze the long-term equilibrium relationship and explore which is the leading indicator using Granger-Causality test. Data is cited from January first ,2016 to June 23,2016 , before the period of Brexit and June 24,2016 to Feburary 28,2017, after the period of Brexit.
The empirical evidence suggests that there is a long-run equilibrium relationship between US Dollar Index、Euro,、Yuan、JPY and GBP during the time period investigated. In the shor-run term, the result of Traditional ECM tests
show there is a unidirectional causal relationship from the Pound to Yuan after Brexit. Moreover, we also find it doesn''t exit causal relationship between US Dollar Index、Euro、Yuan、JPY and GBP whether M-TECM model or Traditional ECM tests during the period investigated.
Before the Brexit, the results of M-TECM Granger-Causality tests show in the long-run term there is an unidirectional causal relationship from US Dollar Index、Euro、JPY and Yuan to GBP. However, we can find the results of traditional ECM model indicate it doesn''t exist unidirectional causal relationship from GBP to Euro、Yuan and JPY.
After the Brexit ,we can find the results of traditional ECM model indicate it doesn''t exist unidirectional causal relationship from GBP to US Dollar Index and Yuan. However, we firstly find the result of TECM model indicate it doesn''t exist unidirectional causal relationship from GBP to US Dollar Index. Secondly, we can find the results of M-TECM model indicate it exists unidirectional causal relationship from GBP to Euro. Thirdly, the result of TECM test show indicates it exists unidirectional causal relationship from GBP to Yuan. Finally, we can find the result of M-TECM Granger-Causality tests show there is a bidirectional causal relationship between GBP and JPY. |
