What do cold-deck and hot-deck imputation techniques accomplish?

Cold-deck and hot-deck imputation techniques are methodologies used to handle missing data within a dataset by filling in those gaps based on existing values from similar records.

The choice that states copying the missing values from similar records accurately describes the function of both imputation techniques. In the hot-deck method, missing values in a given record are filled in with values from other, similar records within the same dataset. This is done quickly and efficiently, often using the most recent or a randomly selected similar case that has available data. Conversely, cold-deck imputation involves sourcing values from an external dataset or a different historical period, where similar records are found to replace the missing data. Both methods draw fundamentally on the principle of utilizing other similar observations to maintain the integrity and distribution of the dataset while addressing gaps.

The other options do not fully capture the essence of cold-deck and hot-deck imputation. Generating new data points from existing records involves creating synthetic data rather than filling in existing gaps, and while removal of rows with missing data is a common approach, it does not utilize available data effectively. Lastly, filling missing values with the mean of the dataset represents a separate, distinctly different imputation technique that does not rely on similar records.