Robust inference with knockoffs

By Rina Foygel Barber, E.J. Candes, Richard J. Samworth. 2018

136 citations recorded by the source corpus.

We consider the variable selection problem, which seeks to identify important variables influencing a response $Y$ out of many candidate features $X_1, \ldots, X_p$. We wish to do so while offering finite-sample guarantees about the fraction of false positives - selected variables $X_j$ that in fact have no effect on $Y$ after the other features are known. When the number of features $p$ is large (perhaps even larger than the sample size $n$), and we have no prior knowledge regarding the type of dependence between $Y$ and $X$, the model-X knockoffs framework nonetheless allows us to select a model with a guaranteed bound on the false discovery rate, as long as the distribution of the feature vector $X=(X_1,\dots,X_p)$ is exactly known. This model selection procedure operates by constructing "knockoff copies'" of each of the $p$ features, which are then used as a control group to ensure that the model selection algorithm is not choosing too many irrelevant features. In this work, we study the practical setting where the distribution of $X$ could only be estimated, rather than known exactly, and the knockoff copies of the $X_j$'s are therefore constructed somewhat incorrectly. Our results, which are free of any modeling assumption whatsoever, show that the resulting model selection procedure incurs an inflation of the false discovery rate that is proportional to our errors in estimating the distribution of each feature $X_j$ conditional on the remaining features $\{X_k:k\neq j\}$. The model-X knockoff framework is therefore robust to errors in the underlying assumptions on the distribution of $X$, making it an effective method for many practical applications, such as genome-wide association studies, where the underlying distribution on the features $X_1,\dots,X_p$ is estimated accurately but not known exactly.

Subfields: Artificial Intelligence, Genetics, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Genetics

In Statistics and Probability

Citations in the corpus

Forward stability and model path selection (2021, neutral): “…not necessarily to identify the subset of signals as is the primary objective with tools like knockoffs (Barber and Candes 2015; Candes et al. 2018; Barber et al. 2018; Janson et al. 2016), pseudovariables (Wu et al. 2007; Hu et al. 2018), or stability selection (SS) (Meinshausen and Buhlmann…”
The conditional permutation test for independence while controlling for confounders (2018, methodology): “Belloni et al. (2014) proposed a variant on this approach, which assumes approximate linear models for both Y and X. Their method regresses both X and Y on Z and then tests for correlation between the two resulting residual vectors; this ‘double regression’ offers superior performance by removing mu”