Skip to main content

Table 1 Model characteristics for Gamma GLM, PLAQR, and Random Forest (RF) models

From: Comparison of statistical and machine learning models for healthcare cost data: a simulation study motivated by Oncology Care Model (OCM) data

 Gamma GLMPLAQRRandom Forest
Distribution assumptionParametricSemi-parametricNonparametric
EstimateMeanQuantileMean
Ability to model skewed outcomeYesYesYes
Non-linear effectNeeds to be specified through model diagnosticsNeeds to be specified (B-spline)Data-driven detection; pre-specification not needed
Interaction effectNeeds to be specified through model diagnosticsNeeds to be specified through model diagnosticsData-driven detection; pre-specification not needed
SoftwareR, SAS, STATARR, SAS