Positive autocorrelation occurs when a data series follows a similar pattern over time, meaning that an increase or decrease in the value of a variable at one point in time is likely to be followed by a similar increase or decrease in future periods. This results in a clustering effect of high and low values.
Negative autocorrelation, on the other hand, indicates that an increase in the value of a variable at one point in time is likely to be followed by a decrease in the next period, and vice versa. This leads to an alternating pattern of high and low values.
The Durbin-Watson statistic can be used to detect autocorrelation. A value close to 2 suggests no autocorrelation; less than 1 implies positive autocorrelation while greater than 3 implies negative autocorrelation.
