# Autoregression Models

Autoregression models are classical or "standard" modeling methods used on time series data (that is, any dataset that changes with time) and can complement the linear regression techniques covered previously. Autoregression models are often used for forecasting in the economics and finance industry as they are useful with univariate time series (where there are no *x* variables other than time) and with very large datasets (such as streaming data or high-frequency sensor data) where the linear algebra operations might run into memory or performance issues on very large datasets. The "auto" part of autoregression refers to the fact that these models leverage correlation of a time series to itself in the past, hence autoregression. In addition, many systems do not have an associated causal model—the time series data is said to be stochastic. An example is stock price data over time. Although many attempts have been made, and continue to...