Time series are data observed over time (either in continuous time or at discrete time periods).
Overview
Time series are data observed over time (either in continuous time or at discrete time periods), usually at recurring intervals such as days, weeks, months, or years. (Series of data irregularly spaced in time that record random events are usually termed "point processes" and are analyzed with methods distinct from those of time series analysis.)
Time series analysis includes trend identification, temporal pattern recognition, spectral analysis, and forecasting future values (or any other values at times without observations).
The salient characteristic of methods of time series analysis (as opposed to more general methods to analyze relationships among data) is accounting for the possibility of serial correlation (also known as autocorrelation and temporal correlation) among the data. Positive serial correlation means successive observations in time tend to be close to one another, whereas negative serial correlation means successive observations tend to oscillate between extremes. Time series analysis also differs from analyses of more general stochastic processes by focusing on the inherent direction of time, creating a potential asymmetry between past and future.
References
This thread contains a list of references on time series:
The following journals are dedicated to researching time series: