A NOTE ON NONSTATIONARY ARMA PROCESSES WITH INFINITE VARIANCE

被引：0

作者：

NASSIUMA, D ^{[1
]}

机构：

[1] UNIV MANITOBA,DEPT STAT,WINNIPEG R3T 2N2,MANITOBA,CANADA

来源：

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS | 1991年 / 20卷 / 5-6期

关键词：

NONSTATIONARY; INFINITE VARIANCE; FORECASTING; STABLE DISTRIBUTIONS; BANACH SPACES;

D O I：

10.1080/03610929108830599

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

We consider the autoregressive moving average (ARMA) process (zeta-t ; t epsilon-T = (0, +/- 1, +/- 2, ...) with nonstationary stable innovations. we give conditions for the existence of a unique solution to the process. we also obtain linear forecasts and the dispersion of the h-steps ahead forecast error.

引用

页码：1821 / 1826

页数：6

共 6 条

[1] LINEAR PREDICTION OF ARMA PROCESSES WITH INFINITE VARIANCE [J].

CLINE, DBH ;

BROCKWELL, PJ .

STOCHASTIC PROCESSES AND THEIR APPLICATIONS, 1985, 19 (02) :281-296

[2]

Feller W., 1966, INTRO PROBABILITY TH, V2

[3] INFINITE VARIANCE AND RESEARCH STRATEGY IN TIME SERIES ANALYSIS [J].

GRANGER, CWJ ;

ORR, D .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1972, 67 (33) :275-&

[4]

NIEMI H, 1983, SCAND J STAT, V10, P11

[5]

PRIESTLY MB, 1988, NONLINEAR NONSTATION

[6] MINIMUM ERROR DISPERSION LINEAR-FILTERING OF SCALAR SYMMETRIC STABLE PROCESSES [J].

STUCK, BW .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1978, 23 (03) :507-509

← 1 →