A stationary process has a joint distribution invariant under a common shift of all time indices.
A weakly stationary process has constant finite mean and covariance depending only on lag.
The autocovariance at lag is , independent of for a weakly stationary process.
The autocorrelation function is when .
The sample autocorrelation function replaces the mean and lagged covariance in by their empirical counterparts.
White noise has constant mean, constant variance, and zero autocovariance at every nonzero lag. Gaussian white noise additionally has jointly Gaussian coordinates and is therefore independent across time.
An autoregressive–moving-average model satisfies
where is the backshift operator, and are finite polynomials, and is white noise.
The backshift operator acts on a time series by and hence .
An autoregressive model expresses the current value as a linear combination of finitely many past values plus white noise.
An autoregressive process of order one satisfies . It is causal and weakly stationary when .
A moving-average model expresses the current value as a finite linear combination of present and past white-noise innovations.
A moving-average process of order one has the form and zero autocovariance beyond lag one.

Articles by others on the same topic (1)

A time series is a sequence of data points recorded or measured at successive points in time, typically at uniform intervals. It is a common method in statistics and various fields, such as finance, economics, environmental science, and engineering, for analyzing trends, patterns, and behaviors of data over time. Key characteristics of time series data include: 1. **Temporal Order**: The data points are ordered chronologically. Each observation has a timestamp, and the order matters.