UZH-Logo

Maintenance Infos

The use of transfer function models, intervention analysis and related time series methods in epidemiology


Helfenstein, Ulrich (1991). The use of transfer function models, intervention analysis and related time series methods in epidemiology. International Journal of Epidemiology, 20(3):808-815.

Abstract

In epidemiology, data often arise in the form of time series e.g. notifications of diseases, entries to a hospital, mortality rates etc. are frequently collected at weekly or monthly intervals. Usual statistical methods assume that the observed data are realizations of independent random variables. However, if data which arise in a time sequence have to be analysed, it is possible that consecutive observations are dependent. In environmental epidemiology, where series such as daily concentrations of pollutants were collected and analysed, it became clear that stochastic dependence of consecutive measurements may be important. A high concentration of a pollutant today e.g. has a certain inertia i.e. a tendency to be high tomorrow as well. Since the early 1970s, time series methods, in particular ARIMA models (autoregressive integrated moving average models) which have the ability to cope with stochastic dependence of consecutive data, have become well established in such fields as industry and economics. Recently, time series methods are of increasing interest in epidemiology. Since these methods are not generally familiar to epidemiologists this article presents their basic concepts in a condensed form. This may encourage readers to consider the methods described and enable them to avoid pitfalls inherent in time series data. In particular, the following topics are discussed: Assessment of relations between time series (transfer function models). Assessment of changes of time series (intervention analysis), forecasting and some related time series methods.

Abstract

In epidemiology, data often arise in the form of time series e.g. notifications of diseases, entries to a hospital, mortality rates etc. are frequently collected at weekly or monthly intervals. Usual statistical methods assume that the observed data are realizations of independent random variables. However, if data which arise in a time sequence have to be analysed, it is possible that consecutive observations are dependent. In environmental epidemiology, where series such as daily concentrations of pollutants were collected and analysed, it became clear that stochastic dependence of consecutive measurements may be important. A high concentration of a pollutant today e.g. has a certain inertia i.e. a tendency to be high tomorrow as well. Since the early 1970s, time series methods, in particular ARIMA models (autoregressive integrated moving average models) which have the ability to cope with stochastic dependence of consecutive data, have become well established in such fields as industry and economics. Recently, time series methods are of increasing interest in epidemiology. Since these methods are not generally familiar to epidemiologists this article presents their basic concepts in a condensed form. This may encourage readers to consider the methods described and enable them to avoid pitfalls inherent in time series data. In particular, the following topics are discussed: Assessment of relations between time series (transfer function models). Assessment of changes of time series (intervention analysis), forecasting and some related time series methods.

Citations

53 citations in Web of Science®
57 citations in Scopus®
Google Scholar™

Altmetrics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:04 Faculty of Medicine > Epidemiology, Biostatistics and Prevention Institute (EBPI)
Dewey Decimal Classification:610 Medicine & health
Language:English
Date:September 1991
Deposited On:24 Jun 2016 14:21
Last Modified:26 Jun 2016 06:55
Publisher:Oxford University Press
ISSN:0300-5771
Publisher DOI:https://doi.org/10.1093/ije/20.3.808
PubMed ID:1955267

Download

Full text not available from this repository.
View at publisher

TrendTerms

TrendTerms displays relevant terms of the abstract of this publication and related documents on a map. The terms and their relations were extracted from ZORA using word statistics. Their timelines are taken from ZORA as well. The bubble size of a term is proportional to the number of documents where the term occurs. Red, orange, yellow and green colors are used for terms that occur in the current document; red indicates high interlinkedness of a term with other terms, orange, yellow and green decreasing interlinkedness. Blue is used for terms that have a relation with the terms in this document, but occur in other documents.
You can navigate and zoom the map. Mouse-hovering a term displays its timeline, clicking it yields the associated documents.

Author Collaborations