What is the term for the correlation in residuals when error terms at successive points in time are related?

The concept of autocorrelation specifically pertains to the correlation of residuals in a time series context, where the error terms at successive points in time are related. When you have autocorrelation, it indicates that the residuals from one time period can be predicted based on the residuals from previous time periods. This can lead to inefficiencies in estimating the coefficients of a regression model because it violates the assumption of independence among error terms, which can result in misleading statistical inferences.

Understanding autocorrelation is crucial when dealing with time series data because it affects the reliability of your model's predictions. If autocorrelation is present, it may suggest that the model is not fully capturing the underlying patterns in the data, or that important variables have been left out, which could otherwise account for the temporal dependencies in the errors.

The other terms mentioned have different meanings. Heteroscedasticity refers to the situation where the variance of the residuals is not constant across all levels of the independent variable, while multicollinearity addresses the issue of high correlation between independent variables, potentially leading to unreliable coefficient estimates. Ignored variance is not a recognized statistical term in the same context as the others, further distinguishing autocorrelation as the correct choice for the description of correlated residuals