When utilizing autoregressive models we are interested in correlation of the signal with a delayed copy of itself. This is known as autocorrelation.

We can calculate the autocorrelation function (ACF) of a signal and a particular lag amount by calculating the Pearson Correlation of the time series and its lagged duplicate equivalent.

In other words, we treat the time series as the X variable, and then take each point in the time series and shift it by whatever lag we’re interested in and use that as the Y variable.

Notes mentioning this note

Here are all the zettels in this zettelkasten, along with their links, visualized as a graph. You may need to zoom and pan around to see something.