Research Topics:

Data Analysis

Content Type:

Glossary

Autocorrelation Definition

The problem of interdependence among successive values of the disturbance term. The problem with autocorrelation concerns the variance of your estimator.

In reference to marketing research, autocorrelation is the statistical relationship between a variable's current value and its past values over a certain period of time, thus determining whether or not the values are related. The intended purpose of autocorrelation is to reveal hidden patterns and relationships within time periods. Determining whether or not autocorrelation exists can lead to more accurate forecasts, improved predictions and enhanced decision-making. Conversely, not addressing autocorrelation can cause problems in forecasts and strategies that can lead to lost resources and opportunities. 

Who relies on autocorrelation?

Market researchers, analysts and economists use autocorrelation to better understand relationships within financial and economic findings, especially those concerning, stock prices, economic indicators and consumer behavior. Autocorrelation also is utilized by investment professionals, financial institutions and policymakers.

Why should I care about autocorrelation?

Professionals and entities in marketing research that seek to interpret findings over time most likely have a knowledge of autocorrelation. That know-how can lead to accurate analysis, enhance predictions and informed decision-making. Not paying attention or autocorrection can lead to disastrous results, among them biased research results, misleading trends and incorrect conclusions.