Heckman imputation models for binary or continuous MNAR outcomes and MAR predictors

Abstract Background Multiple imputation by chained equations (MICE) requires specifying a suitable conditional imputation model for each incomplete variable and then iteratively imputes the missing values. In the presence of missing not at random (MNAR) outcomes, valid statistical inference often re...

Celý popis

Podrobná bibliografie

Hlavní autoři:	Jacques-Emmanuel Galimard, Sylvie Chevret, Emmanuel Curis, Matthieu Resche-Rigon
Médium:	Artigo
Jazyk:	Inglês
Vydáno:	BMC 2018-08-01
Edice:	BMC Medical Research Methodology
Témata:	Heckman’s model Missing data Missing not at random (MNAR) Multiple imputation by chained equation (MICE) Sample selection method
On-line přístup:	http://link.springer.com/article/10.1186/s12874-018-0547-1
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!