How does serial correlation affect observations collected over time?

Created from Britannica articles using AI. AI can make mistakes, so verify using Britannica articles.

Serial correlation, also known as autocorrelation, refers to the correlation between paired values of a statistical variable collected at constant intervals over time.1

In statistics, this phenomenon indicates that observations are not independent of one another. When data points are serially correlated, the value of an observation at a specific time is influenced by or related to the values of preceding observations. This relationship can reveal the degree of periodicity within a dataset, meaning the data may exhibit repeating patterns or trends over time.1

Because standard statistical methods often assume that observations are independent, failing to account for serial correlation can lead to biased estimates or incorrect conclusions regarding the significance of the data.