Description
Statistics and machine-learning workflows can handle datasets with or without missing values. This helps users analyze imperfect real-world data without discarding every incomplete record.
It is a data-analysis extension. Missing-value handling can change conclusions, so methods and assumptions should be documented.