本文以貝氏分析的方法估計非高斯移動平均模式的參數。傳統的quasi likelihood估計法(QMLE)是以Gaussian likelihood 當做估計時的目標函數，而Huang 和 Pawitan (2000)的quasi-likelihood的估計法(H&P)則是以Laplace likelihood當作目標函數。不論所考慮的MA模式是可逆或不可逆，相較於傳統的QMLE及H&P的方法，本研究所提出之方法在所考慮的有限樣本下都有較佳的估計表現。

A method for estimating parameters in non-Gaussian moving average models is proposed based on Bayesian analysis. In the conventional approach, the Gaussian likelihood is used for parameter estimation (QMLE). However, it may not be appropriate when the process is non-invertible. Huang and Pawitan (2000) proposed another likelihood-based estimation method using Laplace likelihood (H&P). Comparing among these three methods, the empirical results show that the Bayesian analysis performs better than QMLE and H&P in terms of smaller root mean square error no matter the MA process is invertible or non-invertible.

Benveniste, A., Goursat, M., and Roget, G. (1980). Robust identification of a nonminimum phase system: blind adjustment of linear equalizer in data communications. IEEE Transactions on Automatic Control, AC-25, 385-398.

Breidt, F.J. and Hsu, N.-J. (2004). Best mean square prediction for moving averages.
forthcoming in Statistica Sinica.

Chien, H.-M., Yang, H.-L., and Chi, C.-Y. (1997). Parametric cumulant base phase estimation of 1-d and 2-d nonminimum phase systems by allpass filtering. IEEE Transactions on Signal Processing, 45, 1742-1762.

Donoho, D. (1981). On minimum entropy deconvolution. In Applied Time Series
Analysis II, (D.F. Findley, ed.), Academic Press, New York, 565-608.

Gelman, A., Carlin, J.S., Stern H.S. and Rubin D.B. (1995). Bayesian Data Analysis, Chapman and Hall, London.

Giannakis G.B. (1987). On the identifiability of non-Gaussian ARMA models using cumulants. IEEE Transactions on Automatic Control, 35, 18-26.

Godfrey, R. and Rocca, F. (1981). Zero memory nonlinear deconvolution. Geophysical Prospecting, 29, 189-228.

Green, P.J. (1995). Reversible jump MCMC computation and Bayesian model determination, Biometrika, 82, 711-732.

Huang, J. and Pawitan, Y. (2000). Quasi-likelihood estimation of non-invertible moving average processes. Scandinavian Journal of Statistics, 27, 689-702.

Lii, K.-S. and Rosenblatt, M. (1982). Deconvolution and estimation of transfer function phase and coefficients for nonGaussian linear processes. Annals of Statistics, 10, 1195-1208.

Lii, K.-S. and Rosenblatt, M. (1992). An approximate maximum likelihood estimation for non-Gaussian non-minimum phase moving average processes. Journal of Multivariate Analysis, 43, 272-299.

Lii, K.-S. and Rosenblatt, M. (1996). Maximum likelihood estimation for
nonGaussian nonminimum phase ARMA sequences. Statistica Sinica, 6, 1-22.

Wiggins, R.A. (1978). Minimum entropy deconvolution. Geoexploration, 16, 21-35.