The Exchange Rate-Interest Differential Relation with Heavy-Tailed Processes

匯率和利率差之關聯性與厚尾數列

王淳玄(Chun-Hsuan Wang) ； 林俊宏(Chun-Hung A. Lin)

《經濟論文叢刊》 31卷4期 (2003/12) Pp. 463-482

https://doi.org/10.6277/ter.2003.314.3

摘要

標準共整合檢定程序中用以評估檢定統計值之分配理論有一重要基本假設，即其誤差項之變異數需為有限。然而，近來有些研究發現，許多金融市場報酬率及匯率變動率可能具有無限變異數誤差之特性。此發現給予我們重新分析實質匯率和實質利率差之共整合關係的主要動機。此研究資料包括加拿大、法國、德國、英國及美國之有關利率、匯率、物價水準等季資料。結果發現，數列中具有厚尾之特性。在以殘差為基礎之共整合檢定中，無論其誤差項基本假設為有限或無限變異數，其結果均拒絕實質匯率和實質利率差之共整合關係。然而，傳統多變量共整合檢定則支持其共整合關係。相對於傳統多變量共整合檢定結果，考慮誤差項變異數為無限之多變量共整合檢定結果發現，其支持證據較為薄弱。

關鍵字

實質匯率 ； 實質利率差 ； 厚尾 ； 共整合

並列摘要

Standard cointegration tests assuming Gaussian processes are not robust to heavy-tails, but some recent studies have suggested many financial market returns and exchange rate returns may be driven by infinite-variance innovations. This is the motivation for re-analyzing the relationship between real exchange rates and real interest differentials (RERI) by allowing for infinite variance processes. Quarterly observations are used for Canada, France, Germany, Japan, UK and US over the recent floating-rate period. We find sufficient evidence of heavy-tails in our data. Accordingly, we re-form residual-based and likelihood-ratio-based cointegration tests using the critical values adjusted for infinite-variance innovations. Our results obtained by the residual-based cointegration tests demonstrate the rejection of RERI link in all cases either under the assumption of Gaussian or α-stable innovations. However, the likelihood-ratio- based cointegration tests which take into consideration heavy-tailed processes provide marginally less supportive evidence than standard likelihood-ratio-based cointegration tests.

並列關鍵字

real exchange rate ； real interest differentials ； heavy tails ； cointegration

參考文獻

Adler, R., Feldman, R., Gallagher, C.(1998).A Practical Guide to Heavy Tailed Data.Boston, MA:Birkhauser.

Akgiray, V., Booth, G. G., Seifert, B.(1988).Distribution properties of Latin American Black-market exchange rates.Journal of International Money and Finance.7

Boothe, P., Glassman, D.(1987).The Statistical Distribution of Exchange Rates: Empirical Evidence and Economic Implication.Journal of International Economics.22

Box, G., Jenkins, G.(1976).Time Series Analysis, Forecasting, and Control.San Francisco, Calif.:Holden Day.

Caner, M.(1998).Tests for cointegration with infinite variance errors.Journal of Econometrics.86