本文主要研究議題為提出雙變量時間序列獨立性檢定方法。
此無母數檢定方法在平穩序列的假設下，
不須事先以適當模型濾化(filtering)，
也不須決定如何選取適當核函數(kernel function)與其平滑參數(smoothing parameter)
來調整統計量中的時期參數，
而直接以兩序列的經驗交叉相關係數(empirical cross correlation function)
與隨時期遞減的時期權數來建構獨立性檢定統計量。
同時，此無母數檢定方法非但檢定程序中所需的主觀設定少外，
其檢定結果也提供有用的訊息以利後續單變量或雙變量模型建模程序。
文中提出檢定不同時期與同時期的時間相關檢定統計量，
兩者的虛無漸進分配為可由樣本大小決定權數參數與自由度參數的卡方分配。

在考慮VARMA模型、具極端值情況雙變量模型、
與尾端相關(tail dependence)模型的模擬分析下，
分別比較本文與Haugh's、Hong's及
Pierre and Roy's等獨立性檢定方法的檢定表現與檢定差異。
透過模擬比較，
本文所提的雙變量時間序列檢定方法除了不受極端值出現而改變原有的檢定表現外，
極端值大小改變時對檢定能力的表現也顯著的優於其他檢定方法，為具穩健性的檢定。
對於可能存在於尾端相關的時間序列模型中，本文提出的檢定方法也優於文獻上的檢定。

\renewcommand{\refname}{}
\begin{thebibliography}{40}
\bibitem{}Ben, M. G., Martinez, E. J. and Yohai, V. J. (1999), Robust estimatiom in
vector autogressive moving-average models. {\it J. Time Ser. Anal.} {\bf 20}, 381-399.
\bibitem{}Blum, J. R., Kiefer, J. and Rosenblatt, M. (1961). Distribution free tests of indenpendence
based on the sample distru function. {\it Ann. Math. Statist.} {\bf 32}, 485-498.
\bibitem{}Box, G. E. P., Jenkins, G. M. and Reinsel, G. C. (1994).
{\it Time Series Analysis. Forecasting and Control.} 3rd edition. Prentice Hall, Englewood Cliffs, NJ.
\bibitem{}Brockwell, P. J. and Davis, R. A. (1996). {\it Introduction to Time Series and Forecasing.}
Springer, New York.
\bibitem{}Bustos, O. H. and Yohai, V. J. (1986). Robust estimates for ARMA models.
{\it J. Amer. Statist. Assoc.} {\bf 81}, 155-168.
\bibitem{}Chan, N. H. and Tran, L. T. (1992). Nonparametric tests for serial dependence.
{\it J. Time Ser. Anal}. {\bf 13}, 102-113.
\bibitem{}Chitturi, R. V., (1970). Distribution of residual autocorrelations in multiple
autoregressive schemes. {\it J. Am. Statist. Assoc.} {\bf 69}, 928-934.
\bibitem{}Coles, S.G. (2001), {\it An Introduction to Statistical Modelling of Extreme Values}.
Springer Series in Statistical.
\bibitem{}Coles, S.G. and Tawn, J.A. (1991), Modelling extreme multivariate events.
{\it Journal of the Royal Statistical Society B,} {\bf 53}, 377-392.
\bibitem{}Coles, S.G. and Tawn, J.A. (1994), Statistical methods for multivariate extremes:
An application to structural design (with discussion). {\it Applied Statistics,} {\bf 43}, 1-48.
\bibitem{}Deheuvels, P. (1981). An asymptotic decomposition for multivariate distribution-free
tests of indenpendence. {\it J. Mult. Anal.} {\bf 11}, 102-113.
\bibitem{}Derby, L. and Martin, R. D. (1979). Robust estimation of the first-order autoregression.
{\it Amer. Statist.} {\bf 42}, 285-287.
\bibitem{}El Himdi, K. and Roy, R. (1997). Tests for noncorrelation of two multivariate ARMA time series.
{\it Canad. J. Statist.} {\bf 25}, 233-256.
\bibitem{}Gewekw, J. (1981). A comparison of tests of indenpendence of two covariance stationary time series.
{\it J. Am. Sataist. Assoc.} {\bf 76}, 363-373.
\bibitem{}Goodman, L. A. and Grunfeld, Y. (1961). Some nonparametric tests for comovements between time series.
{\it J. Am. Statist. Assoc.} {\bf 56}, 11-26.
\bibitem{}Granger, C. W. J. (1980). Testingfor causality: a personal viewpoint.
{\it J. Economic Dynamics and Control} {\bf 2}, 329-352.
\bibitem{}Haan, L. de and Ronde, J. de (1998). Sea and wind:multivariate extremes
at work. {\it Extremes} {\bf 1}, 7-45.
\bibitem{}Hallin, M., Jurevckova, J. Picek, J. and Zahaf, T. (1999).
Nonparametric tests of independence of two autoregressive time series based on autoregression rank scores.
{\it J. Statist. Plann. Inference} {\bf 75}, 319-352.
\bibitem{}Haugh, L. D. (1976). Checking the independence of two covariance-stationary
time series: a univariate residual cross-correlation approach.
{\it J. Amer. Statist. Assoc.} {\bf 71}, 378-385.
\bibitem{}Haugh, L. D. and Box, G. E. P. (1977). Identification of dynamic regression(distributed lag)
models connecting two time series. {\it J. Am. Statist. Assoc.} {\bf 72}, 121-130.
\bibitem{}Hoeffding, W. (1948). A non-parametric test of indenpendence. {\it Ann. Math. Statist.}
{\bf 19}, 546-557.
\bibitem{}Hong, Y. (1996). Testing for indenpendence between two stationary time series.
{\it Biometrika} {\bf 83}, 615-625.
\bibitem{}Hong, Y. (1997). One-sided testing for conditional heteroscedasticity in time series models.
{\it J. Time Ser. Anal.} {\bf 18}, 253-277.
\bibitem{}Hong, Y. (2000). Generalized spectral tests for serial dependence.
{\it J. R. Statist. Soc. B} {\bf 62}, 557-574.
\bibitem{}Hosking, J. (1980). The multivariate portmanteau statistic.
{\it J. Amer. Statist. Assoc.} {\bf 75}, 602-608.
\bibitem{}Huber, P. J. (1981). {\it Robust Statistics,} Wiley, New York.
\bibitem{}Joe, H. (1997), {\it Multivariate Models and Dependence Concepts.}
Monographs on Statistics and Applied Probability, {\bf 73}, Chapman \& Hall, London.
\bibitem{}Joe, H. , Smith, R.L. and Weissman, I. (1992), Bivariate threshold methods for extremes.
{\it Journal of the Royal Statistical Society B,} {\bf 54}, 171-183.
\bibitem{}Koch, P. D. and Yang, S. S. (1986). A method for testing the indenpence of two time series
that accounts for a potential pattern in the cross-correlation function. {\it J. Amer. Statist. Assoc.}
{\bf 81}, 533-544.
\bibitem{}Ledford, A.W. and Tawn, J.A. (1997), Modelling dependence within joint tail regions.
{\it Journal of the Royal Statistical Society B}, {\bf 59}, 475-499.
\bibitem{}Li, W. K. (1988). A goodness-of-fit test in robust time series modelling.
{\it Biometrika} {\bf 75}, 355-361.
\bibitem{}Li, W. K. and Hui, Y. V. (1994). Robust residual cross correlation tests for
lagged relations in time series. {\it J. Statist. Comput. Simulation} {\bf 49}, 103-109.
\bibitem{}Ljung, G. M. and Box, G. E. P. (1978) On a measure of lack of fit in time series models.
{\it Biometrika} {\bf 65}, 297-303.
\bibitem{}McLeod, A. I. (1979). Distribution of the residual cross-crorelation in univariate
ARMA time series models. {\it J. Amer. Statist. Assoc.} {\bf 74}, 849-855.
\bibitem{}Patton, A.J. (2001), Modelling time-varying exchange rate dependence using the conditional copula.
Manuscript, University of California.
\bibitem{}Pham, D. T., Roy, R. and Cedras, L. (2003). Tests for non-correlation of two cointegrated
ARMA time series.{\it J. Time Ser. Anal., } in press.
\bibitem{}Pierce, A. (1977). Lack of dependence among economic variables.
{\it J. Am. Statist. Assoc.} {\bf 72}, 11-22.
\bibitem{}Pierce, D. A. and Haugh, L. D. (1979). Causality in temporal systems.
{\it J. Economic} {\bf 5}, 265-293.
\bibitem{}Pierre, D. and Roy, R. (2003). Robust tests for independence of two time series.
{\it Statistica Sinica.} {\bf 13}, 827-852.
\bibitem{}Priestley, M. B. (1981). {\it Spectral Analysis and Time Series}, {\bf 1, 2}.
London:Academic Press.
\bibitem{}Rousseeuw, P. J. and Leroy, A. M. (1987) {\it Robust Regression and Outlier Detection.}
Wiley, New York.
\bibitem{}Satterthwaite, F. E. (1941). Synthesis of variance. {\it Psychmetrica}, {\bf 6}. 309-316.
\bibitem{}Serfling, R. (1980). {\it Approximation Theorems of Mathematical Statistics.}
New York, Wiley.
\bibitem{}Shao, J. and Tu, D. (1995). {\it The Jacknife and Bootstrap}.New York, Springer Verlag.
\bibitem{}Shorack, G. R. and Wellner, J. A. (1986). {\it Empirical Processes with Applications to Statistics.}
New York, Wiley.
\bibitem{}Skaug, H. J. and Tjostheim, D. (1993). A nonparametric test of serial independence based
on the empirical distribution function. {\it Biometrika} {\bf 80}, 591-602.
\bibitem{}Solomon, Herbert, and Stephens, Michael A. (1977). Distribution of a sum of
weighted chi-square variables. {\it J. Am. Statist. Assoc.}, {\bf 72}. 881-885.
\bibitem{}Stute, W. and Schumann, G. (1980). The general Glivenko-Cantelli theorem for stationary
sequences of random observations. {\it Scand. J. Statist.} {\bf 7}, 102-104.
\bibitem{}Tsay, R. S. (2001). {\it Analysis of Financial Time Series}.New York, Wiley.
\bibitem{}Tsay, R. S. and Tiao, G. C. (1995). Asymptotic properties of multivariate nonstationary
processes with applications to autoregressions. {\it Ann. Statist.} {\bf 18}, 220-250.
\bibitem{}Wand, M. P. and Jones, M. C. (1995). {\it Kerneel Smoothing}, London, Chapman and Hall.
\bibitem{}Zhang, J. T. (2005). Approximate and asymptotic distribution of chi-squared-type
mixtures with applications. {\it J. Am. Statist. Assco.} {\bf 100}, 273-285.
\bibitem{}Zhang, Z. and Smith, R.L. (2001), Modelling financial time series data as moving maxima processes.
Manuscript, University of North Carolina.
\end{thebibliography}