Multiple imputation using nearest neighbor methods

Missing values are a major problem in medical research. As the complete case analysis dis-cards useful information, estimation and inference may suffer strongly. Multiple imputa-tion has been shown to be a useful strategy to handle missing data problems and account for the uncertainty of imputation. In the presence of high-dimensional data (p >> n), the missing values raise even more serious problems as the existing software packages tend to fail. We present multiple imputation methods based on nearest neigh-bors. The distances are computed using the information of correlation among the target and candidate predictors. Thus only the relevant predictors contribute for computing dis-tances. The method successfully imputes missing values also in high-dimensional settings. Using a variety of simulated data with MCAR and MAR missing patterns, the proposed algo-rithm is compared to existing methods. Various measures are used to compare the perfor-mance of methods, including MSE for imputation, MSE of estimated regression coefficients, their standard errors, confidence intervals, and their coverage probabilities. The simulation results, for both cases n < p and n > p, show that the sequential imputation using weighted nearest neighbors can be successfully applied to a wide range of data settings and outper-forms or is close to the best when compared to existing methods. (c) 2021 Elsevier Inc. All rights reserved.

Automated summary

The article discusses the use of multiple imputation methods to address missing values in medical research, particularly in high-dimensional data settings. The proposed method based on nearest neighbors successfully imputes missing values and performs well in simulated data comparisons.