Donor limited hot deck imputation: effects on parameter estimation

Methods for dealing with missing data in the context of large surveys or data mining projects are limited by the computational complexity that they may exhibit. Hot deck imputation methods are computationally simple, yet effective for creating complete data sets from which correct inferences may be drawn. All hot deck methods draw values for the imputation of missing values from the data matrix that will later be analyzed. The object, from which these available values are taken for imputation within another, is called the donor. This duplication of values may lead to the problem that using any donor “too often” will induce incorrect estimates. To mitigate this dilemma some hot deck methods limit the amount of times any one donor may be selected. This study answers which conditions influence whether or not any such limitation is sensible for six different hot deck methods. In addition, five factors that influence the strength of any such advantage are identified and possibilities for further research are discussed.