Imputation by interpolation employs nonlinear curve fitting techniques to predict missing values based other parts of the data that are available.

This method is based on interpolation of the time series of data with missing values, using the intrinsic pattern in the time series, without the need for other data sources.

This is the preferred imputation method if there are no other data sources that are sufficiently related to the data with missing values, if the data with missing values varies smoothly, and if the maximum number of subsequent missing values is not very high.