2014/11/05 by Benjamin Hofner, Hofner, Benjamin, Luigi Boccuto +3 · 2 citations
Biochemistry, Genetics and Molecular Biology · Mathematics · #Gene expression and cancer classification #Statistical Methods and Inference #Gene Regulatory Network Analysis
paper · pdf · doi:10.48550/arxiv.1411.1285
Modern biotechnologies often result in high-dimensional data sets with much\nmore variables than observations (n \≪ p). These data sets pose new\nchallenges to statistical analysis: Variable selection becomes one of the most\nimportant tasks in this setting. We assess the recently proposed flexible\nframework for variable selection called stability selection. By the use of\nresampling procedures, stability selection adds a finite sample error control\nto high-dimensional variable selection procedures such as Lasso or boosting. We\nconsider the combination of boosting and stability selection and present\nresults from a detailed simulation study that provides insights into the\nusefulness of this combination. Limitations are discussed and guidance on the\nspecification and tuning of stability selection is given. The interpretation of\nthe used error bounds is elaborated and insights for practical data analysis\nare given. The results will be used to detect differentially expressed\nphenotype measurements in patients with autism spectrum disorders. All methods\nare implemented in the freely available R package stabs.\n