Absolute error in MSE estimation for individual coefficients for different types of Plasmode simulation compared to parametric simulation under assumption of true data generating process and outcome generating model.
figure
posted on 2024-05-15, 17:34 authored by Marieke Stolte, Nicholas Schreck, Alla Slynko, Maral Saadati, Axel Benner, Jörg Rahnenführer, Andrea Bommert(A) p = 2, n = 100, β = (1, 1, 1)T, σ = 0.3, Cor(Xi, Xj) = 0.2 ∀i ≠ j. (B) p = 2, n = 100, β = (1, 1, 1)T, σ = 3, Cor(Xi, Xj) = 0.2 ∀i ≠ j.
(TIF)
History
Usage metrics
Categories
Keywords
small sampling proportionresampling strategy usedmean squared errorleast squares estimatorcomputer experiments usingresampling feature dataxlink "> simulationneutral simulation studiestrue underlying dataplasmode simulation studiestrue ogm affectparametric simulation differplasmode simulationrealistic datasimulation typesimulation studyparametric simulationplasmode simulationsplasmode influencestrue dgpusually unknowntarget variablesuitable methodstatistical methodsrandom numberspractitioners confrontedoften claimedmany factorslinear regressionlife datasetgreat interestexactly reflectdesign faircrucial toolcompare parametric
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC