A NOTE ON LIKELIHOOD ESTIMATION OF MISSING VALUES IN TIME-SERIES

被引：10

作者：

PENA, D ^{[1
]}

TIAO, GC ^{[1
]}

机构：

[1] UNIV CHICAGO,GRAD SCH BUSINESS,CHICAGO,IL 60637

来源：

AMERICAN STATISTICIAN | 1991年 / 45卷 / 03期

关键词：

ARIMA MODELS; INTERPOLATION; MEAN SQUARED ERROR;

D O I：

10.2307/2684292

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：

Missing values in time series can be treated as unknown parameters and estimated by maximum likelihood or as random variables and predicted by the expectation of the unknown values given the data. The difference between these two procedures is illustrated by an example. It is argued that the second procedure is, in general, more relevant for estimating missing values in times series.

引用

页码：212 / 213

页数：2