|
Abstract:
|
In many applications, it is of interest to identify a parsimonious set of features, or panel, from multiple candidates that achieves a desired level of performance in predicting a response. This task is often complicated in practice by missing data arising from the sampling design or other random mechanisms. Most recent work on variable selection in missing data contexts relies in some part on a finite-dimensional statistical model (e.g., a generalized or penalized linear model). If this model is misspecified, the final panel may not be fully scientifically relevant and may have suboptimal classification performance. To address this limitation, we propose several nonparametric variable selection algorithms combined with multiple imputation to develop flexible panels in the presence of missing-at-random data, and outline strategies that achieve control of commonly used error rates. Through simulations, we show that our proposals achieve more robust performance than existing approaches, and we develop biomarker panels for separating pancreatic cysts with differing malignancy potential in a setting where complicated missingness arose due to limited specimen quantities.
|
