Comparison of Random Forest and Stepwise Regression for Variable Selection Using Low Prevalence Predictors: A case Study in Paediatric Sepsis.
Introduction: Variable selection is a common technique to identify the most predictive variables from a pool of candidate predictors. Low prevalence predictors (LPPs) are frequently found in clinical data, yet few studies have explored their impact on model performance during variable selection. Thi...
| Publicado en: | Maternal & Child Health Journal Vol. 29; no. 5; pp. 604 - 614 |
|---|---|
| Autores principales: | , , , , , , |
| Formato: | research tables/charts Journal Article |
| Publicado: |
Springer Nature
May2025
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=185351226&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 185351226 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 10927875 N9J jtl: Maternal & Child Health Journal issn: 10927875 maglogo: N pubinfo: dt: May2025 vid: 29 iid: 5 pid: 237 pub: Springer Nature place: New York, New York artinfo: ui: 185351226 182231423 185351226 185351226 10.1007/s10995-025-04038-1 185351226 ppf: 604 ppct: 10 formats: fmt: – @attributes: type: T – @attributes: type: P tig: atl: Comparison of Random Forest and Stepwise Regression for Variable Selection Using Low Prevalence Predictors: A case Study in Paediatric Sepsis. aug: au: Gilholm, Patricia Lister, Paula Irwin, Adam Harley, Amanda Raman, Sainath Schlapbach, Luregn J Gibbons, Kristen S affil: https://ror.org/00rqy9422 Children's Intensive Care Research Program, Child Health Research Centre, The University of Queensland, Brisbane, QLD, Australia sug: subj: Neonatal Sepsis Epidemiology Random Forest Step-Wise Multiple Regression Algorithms Models, Statistical Variable Health Screening Human Australia Simulations Prevalence ROC Curve Predictive Value of Tests Prospective Studies Child Confidence Intervals Odds Ratio Neonatal Sepsis Risk Factors Funding Source Child: 6-12 years ab: Introduction: Variable selection is a common technique to identify the most predictive variables from a pool of candidate predictors. Low prevalence predictors (LPPs) are frequently found in clinical data, yet few studies have explored their impact on model performance during variable selection. This study compared the Random Forest (RF) algorithm and stepwise regression (SWR) for variable selection using data from a paediatric sepsis screening tool, where 18 out of 32 predictors had a prevalence < 10%. Methods: Variable selection using RF was compared to forward and backward SWR. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), and the variables retained. Additionally, a simulation study assessed how increasing the prevalence of the predictors impacted the variable selection results. Results: The best fitting RF and SWR models retained were 22, and 17 predictors, respectively, with 14 and 10 predictors having a prevalence < 10%. Both the RF and SWR models had similar predictive performance (RF: AUC [95% Confidence Interval] 0.79 [0.77, 0.81], LR: 0.80 [0.78, 0.82]). The simulation study revealed differences for both RF and SWR models in variable importance rankings and predictor selection with increasing prevalence thresholds, particularly for moderately and strongly associated predictors. Discussion: The RF algorithm retained a number of very low prevalence predictors compared to SWR. However, the predictive performance of both models were comparable, demonstrating that when applied correctly and the number of candidate predictors is small, both methods are suitable for variable selection when using low prevalence predictors. Significance: What is Already Known on this Subject?: Low prevalence predictors (LPPs) are common in clinical data and require special consideration during the variable selection process. Random Forest (RF) is an efficient and increasingly popular algorithm for variable selection. Numerous approaches have been developed using RF for variable selection, which take advantage of the variable importance measures produced by the algorithm. What this Study adds?: This study evaluates the performance of RF for variable selection when using LPPs. The RF and stepwise regression had comparable predictive performance; however, the RF was able to account for complex interactions and was less sensitive to the prevalence of the variables. pubtype: Academic Journal doctype: research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|