Published August 17, 2021 | Version 1
Software Open

R code to perform variable ranking with the Cox model in presence of highly correlated features

  • 1. University of Padova

Description

The R function ranking_cox_bw.R ranks the input features of a Cox model based on their predictive ability, suitably managing the presence of highly correlated numerical features. The function implements the method based on recursive feature elimination and Borda count presented in Vettoretti M. and Di Camillo B., A Variable Ranking Method for Machine Learning Models with Correlated Features: In-Silico Validation and Application for Diabetes Prediction. Appl. Sci. 2021, 11(16), 7740; https://doi.org/10.3390/app11167740

For the function documentation see the file ranking_cox_bw.Rd, while the R markdown file test_ranking_cox_bw_simulated_data.Rmd can be used to test the function with simulated data. 

 

Files

Files (25.3 kB)

Name Size Download all
md5:26ae6314fcec53e83417658b8785f749
9.9 kB Download
md5:cc8eaec5ab76166cd54559c280f32ce5
3.7 kB Download
md5:598f530bf938d90ad1b756411df6f66d
11.6 kB Download

Additional details

Funding

PULSE – Participatory Urban Living for Sustainable Environments 727816
European Commission

References

  • Vettoretti M. and Di Camillo B., A Variable Ranking Method for Machine Learning Models with Correlated Features: In-Silico Validation and Application for Diabetes Prediction. Appl. Sci. 2021, 11(16), 7740