Asymptotic results for spatial ARMA models. (English) Zbl 1093.62083

Illig, Aude; Truong-Van, Benoît

Commun. Stat., Theory Methods 35, No. 4, 671-688 (2006).

Summary: Causal quadrantal-type spatial ARMA\((p, q)\) models with independent and identically distributed innovations are considered. In order to select the orders (\(p, q\)) of these models and estimate their autoregressive parameters, estimators of the autoregressive coefficients, derived from the extended Yule-Walker equations are defined. Consistency and asymptotic normality are obtained for these estimators. Then, spatial ARMA model identification is considered and a simulation study is given.

MSC:

62M10	Time series, auto-correlation, regression, etc. in statistics (GARCH)
62M30	Inference from spatial processes
62F12	Asymptotic properties of parametric estimators

Keywords:

asymptotic properties; causality; estimation; order selection; spatial autoregressive moving-average models

References:

[1]	Brockwell P. J., Time Series: Theory and Methods., 2. ed. (1987) · Zbl 0604.62083
[2]	DOI: 10.1111/j.1467-9892.1991.tb00077.x · Zbl 0729.62081 · doi:10.1111/j.1467-9892.1991.tb00077.x
[3]	Choi B., ARMA Model Identification (1992) · Zbl 0754.62071
[4]	DOI: 10.1080/03610920008832574 · Zbl 0991.62076 · doi:10.1080/03610920008832574
[5]	DOI: 10.1016/0167-7152(94)90052-3 · Zbl 0806.62077 · doi:10.1016/0167-7152(94)90052-3
[6]	Guyon X., Random Fields on a Network (1995) · Zbl 0839.60003
[7]	DOI: 10.1093/biomet/80.1.242 · Zbl 0769.62065 · doi:10.1093/biomet/80.1.242
[8]	Huang D. W., Sci. China Ser. A 35 pp 413– (1992)
[9]	DOI: 10.1111/j.1467-842X.1992.tb01066.x · Zbl 0776.62074 · doi:10.1111/j.1467-842X.1992.tb01066.x
[10]	DOI: 10.1080/03610929508831600 · Zbl 0937.62641 · doi:10.1080/03610929508831600
[11]	DOI: 10.2307/1426722 · Zbl 0383.62060 · doi:10.2307/1426722
[12]	DOI: 10.2307/1426619 · Zbl 0525.62084 · doi:10.2307/1426619
[13]	DOI: 10.2307/2287515 · Zbl 0471.62088 · doi:10.2307/2287515