CONFIDENCE-INTERVALS FOR AUTOCORRELATIONS BASED ON CYCLIC SAMPLES

被引：19

作者：

CLAYTON, MK ^{[1
]}

HUDELSON, BD ^{[1
]}

机构：

[1] UNIV WISCONSIN,DEPT PLANT PATHOL,MADISON,WI 53706

来源：

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION | 1995年 / 90卷 / 430期

关键词：

DIFFERENCE COVER; MISSING DATA; SERIALLY CORRELATED DATA; TIME SERIES; UNEQUALLY SPACED OBSERVATIONS;

D O I：

10.2307/2291088

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：

Clinger and Van Ness have shown how to sample a stochastic process in a periodic manner such that the autocorrelation function for that process can be estimated at all lags. We derive approximate confidence intervals for these estimates, focusing on low-order autoregressive moving average processes. Our methods are illustrated with an example from plant disease epidemiology.

引用

页码：753 / 757

页数：5

共 15 条

[1]

[Anonymous], TIME SERIES ANAL IRR

[2]

[Anonymous], 1990, TIME SERIES ANAL UNI

[3]

BARTLETT MS, 1946, J ROY STAT SOC B, V8, P27

[4]

BOX GEP, 1994, TIME SERIES ANAL