Time Series

Abstract

This book is an introductory account of statistical methods for analysing time-series data, with a focus on the application of the methods to biomedical and population health research. In its simplest form, a time series is a single set of observations measured at equally spaced times. Extensions to this basic format that are particularly relevant in biomedical and population health settings include measurements at irregularly spaced times and independently replicated time series, also called longitudinal or repeated measurement data. The book starts by presenting a set of examples that we use in later chapters to show how the methods can be used to answer specific research questions. Later chapters cover simple descriptive methods for the display and exploratory analysis of time series data without any specific research objective; methods for discrete-time, continuous-time, and replicated time series; and spectral analysis and bivariate time series analysis. Three appendices summarize mathematical and statistical prerequisites. We have aimed to write a book that can be used either as a textbook for a masters-level course in biostatistics or as a manual for applied statisticians and epidemiologists.