Missing Values in Vector Time Series.

We consider the problem of estimating missing values in vector time series. We give a method of calculating the minimum least squares estimate and a second method which uses the best linear combination of the forward and backward predictors. We derive the estimators for some simple models. We also give a method for calculatingthe exact likelihood of Gaussian linear process with missing observations using the innovations algorithm.

MoreLess

Year of publication:	1997
Authors:	Mitchell, H.
Institutions:	School of Economics, Finance and Marketing, RMIT University
Subject:	TIME SERIES \| ESTIMATOR

More details

Series:	Melbourne - Centre in Finance.
Type of publication:	Book / Working Paper
Notes:	13 pages
Classification:	C32 - Time-Series Models ; C10 - Econometric and Statistical Methods: General. General ; C13 - Estimation
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005783581