Methods & evidenceSelecting high-dimensional predictors via automated algorithms reduces residual variance without allowing researcher discretion to bias covariate choices.In randomized trials evaluating public health interventions across thousands of regions, outcomes are influenced by many baseline demographic and epidemiological variables. Including too many collinear variables causes variance inflation, while omitting key predictors reduces statistical precision. Machine learning selection identifies optimal baseline covariates to maximize statistical precision without introducing researcher bias.
RCT