Multicollinearity applied stepwise stochastic imputation: a large dataset imputation through correlation-based regression

Abstract This paper presents a stochastic imputation approach for large datasets using correlation selection methodology when preferred commercial packages struggle to iterate due numerical problems. A variable range-based guard rail modification is proposed that benefits the convergence rate of data elements while simultaneously providing increased confidence in plausibility imputations. country conflict dataset motivates search impute missing values well over common threshold 20% missingness. The Multicollinearity Applied Stepwise Stochastic (MASS-impute) capitalizes on between variables within and uses model residuals estimate unknown values. Examination provides insight toward choosing linear or nonlinear modeling terms. Tailorable tolerances exploit residual information fit each element. evaluation includes observing computation time, fit, comparison known replaced created through imputation. Overall, useable defendable results imputing dataset.