English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Imputation procedures in biomedical fields have turned into statisticalpractice, since further analyses can be conducted ignoring the former presenceof missing values. In particular, non-parametric imputation schemes like therandom forest or a combination with the stochastic gradient boosting have shownfavorable imputation performance compared to the more traditionally used MICEprocedure. However, their effect on valid statistical inference has not beenanalyzed so far. This paper closes this gap by investigating their validity forinferring mean differences in incompletely observed pairs while opposing themto a recent approach that only works with the given observations at hand. Ourfindings indicate that machine learning schemes for (multiply) imputing missingvalues may inflate type-I-error or result in comparably low power in small tomoderate matched pairs, even after modifying the test statistics using Rubin'smultiple imputation rule. In addition to an extensive simulation study, anillustrative data example from a breast cancer gene study has been considered.

A cautionary tale on using imputation methods for inference in matched pairs design