Correlation is the act of comparing two things to one another, and exploring their relationship. When measured on a `-1`

to `1`

scale, the former implies ‘perfect negative correlation’ (when one goes up, the other must go down), whilst the latter implies they are perfectly in lockstep (an increase in one by *x* amount is accompanied by an increase in the other of an equal amount *x*).

Autocorrelation is a measure of the degree of similarity between any time series and a lagged or offset version of itself over successive time intervals.

Instead of comparing one time series to another, the same time series is used twice, offset by a lag. As a result, autocorrelation is sometimes called “lagged correlation” or “serial correlation”.

Like correlation, autocorrelation is measured on a -1 (perfect negative correlation) to 1 (perfect correlation) scale.

**Autocorrelation is used to measure the relationship between the current value of a variable and its past value(s).** High levels of autocorrelation indicate that a variable may not be random.