midTime Series
What is autocorrelation in time series analysis?
Updated May 15, 2026
Short answer
Autocorrelation measures the relationship between a time series and its lagged values.
Deep explanation
Autocorrelation quantifies how current values depend on past values. It is computed using correlation between the series and its lagged versions. High autocorrelation indicates strong temporal dependence, which is essential for forecasting models like ARIMA.
Real-world example
Temperature today is highly correlated with temperature from previous days.
Common mistakes
- Ignoring autocorrelation when selecting forecasting models.
Follow-up questions
- What is partial autocorrelation?
- Why is autocorrelation important?