Number of the records: 1  

Causal mediation analysis with double machine learning

  1. TitleCausal mediation analysis with double machine learning
    Author infoHelmut Farbmacher ... [et al.]
    Author Farbmacher Helmut (20%)
    Co-authors Huber Martin (20%)
    Lafférs Lukáš 1986- (20%) UMBFP10 - Katedra matematiky
    Langen Henrika (20%)
    Spindler Martin (20%)
    Source document The Econometrics Journal. Vol. 25, no. 2 (2022), pp. 277-300. - Londýn : Royal Economic Society, 2022
    Keywords matematické metódy - mathematical methods   ekonomika - economics   strojové učenie - machine learning   analýza kauzálneho sprostredkovania - causal mediation analysis  
    Form. Descr.články - journal articles
    LanguageEnglish
    CountryGreat Britian
    AnnotationThis paper combines causal mediation analysis with double machine learning for a data-driven control of observed confounders in a high-dimensional setting. The average indirect effect of a binary treatment and the unmediated direct effect are estimated based on efficient score functions, which are robust with respect to misspecifications of the outcome, mediator, and treatment models. This property is key for selecting these models by double machine learning, which is combined with data splitting to prevent overfitting. We demonstrate that the effect estimators are asymptotically normal and n−1/2-consistent under specific regularity conditions and investigate the finite sample properties of the suggested methods in a simulation study when considering lasso as machine learner. We also provide an empirical application to the US National Longitudinal Survey of Youth, assessing the indirect effect of health insurance coverage on general health operating via routine checkups as mediator, as well as the direct effect.
    URLLink na plný text
    Public work category ADC
    No. of Archival Copy51676
    Catal.org.BB301 - Univerzitná knižnica Univerzity Mateja Bela v Banskej Bystrici
    Databasexpca - PUBLIKAČNÁ ČINNOSŤ
    ReferencesPERIODIKÁ-Súborný záznam periodika
Number of the records: 1  

  This site uses cookies to make them easier to browse. Learn more about how we use cookies.