| Summary: | In statistical data analysis, penalized regression is considered an attractive approach<br/>for its ability of simultaneous variable selection and parameter estimation. Although<br/>penalized regression methods have shown many advantages in variable selection and<br/>outcome prediction over other approaches for high-dimensional data, there is a relative<br/>paucity of the literature on their applications to hypothesis testing, e.g. in genetic<br/>association analysis. In this study, we apply several new penalized regression methods<br/>with a novel penalty, called Truncated L1-penalty (TLP) (Shen et al. 2012), for<br/>either variable selection, or both variable selection and parameter grouping, in a dataadaptive<br/>way to test for association between a quantitative trait and a group of rare<br/>variants. The performance of the new methods are compared with some existing tests,<br/>including some recently proposed global tests and penalized regression-based methods,<br/>via simulations and an application to the real sequence data of the Genetic Analysis<br/>Workshop 17 (GAW17). Although our proposed penalized methods can improve over<br/>some existing penalized methods, often they do not outperform some existing global<br/>association tests. Some possible problems with utilizing penalized regression methods<br/>in genetic hypothesis testing are discussed. Given the capability of penalized regression<br/>in selecting causal variants and its sometimes promising performance, further studies<br/>are warranted.
|
|---|