U.S. flag

An official website of the United States government

Skip Header


Modeling Time Series Subject to Sampling Error

Written by:
RR89-01

Introduction

Papers by Scott and Smith (1974), and Scott, Smith, and Jones (1977) suggested the use of signal extraction results from time series analysis to improve estimates in periodic surveys. Given models for the true unobserved time series (population quantities) and the sampling errors, these results produce estimates of the population quantities that have minimum mean squared error among estimates that are linear functions of the observed time series of survey estimates. To apply these results in practice one must model the time series structure of both the population quantities and the sampling errors. This presents certain difficulties, which have impeded the adoption of signal extraction techniques by government agencies doing periodic surveys. Research efforts have expanded in recent years to attempt to address some of the problems involved. See, e.g., Hausman and Watson (1985), Miazaki (1985), Rao, Srinath, and Quenneville (1986), Bell and Hillmer (1987), Tam (1987), and Binder and Dick (1986).

Page Last Revised - October 28, 2021
Is this page helpful?
Thumbs Up Image Yes Thumbs Down Image No
NO THANKS
255 characters maximum 255 characters maximum reached
Thank you for your feedback.
Comments or suggestions?

Top

Back to Header