Hierarchical testing of variable importance

By Nicolai Meinshausen. 2008

177 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/932c533c94567e9e283f13cb70be71c10b57ca34

How researchers see this paper

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

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Group bound: confidence intervals for groups of variables in sparse high dimensional regression without assumptions on the design (2013, methodology): “One could also provide a family-wise error control by testing at level α/p · |G| at each cluster G [Meinshausen, 2008], which would amount to a Bonferroni-style correction at the level of individual nodes and the usual level α at the root node but we simply test all groups at the same level α here w”
Simultaneous Inference for High-Dimensional Linear Models (2016, methodology): “Mandozzi and B¨uhlmann (2014) extended the hierarchical testing method in Meinshausen (2008) to the high-dimensional setting, which is able to detect the smallest groups of variables and asymptotically control the family-wise error rate (FWER).”
The Sequential Rejection Principle of Familywise Error Control (2010, methodology): “…multiple testing procedures, such as the exact resampling-based method of Romano and Wolf (2005), as well as recent methods for multiple testing in graphs of logically related hypotheses [Goeman and Mansmann (2008), Meinshausen (2008)], can also be viewed as sequentially rejective procedures.”
High-Dimensional Statistics with a View Toward Applications in Biology (2014, background): “A hierarchical approach where statistical tests are pursued in a top-down fashion from large groups of correlated test-statistics to smaller groups and individual hypotheses is presented in Meinshausen (2008): it is an interesting route to deal with the problem of very high multiplicity in testing.”
Only closed testing procedures are admissible for controlling false discovery proportions (2019, methodology): “Examples of such methods include Meinshausen (2008), Goeman and Mansmann (2008), Goeman and Finos (2012), Meijer and Goeman (2015b), Meijer et al. (2015), and Meijer and Goeman (2015a).”
A simple new approach to variable selection in regression, with application to genetic fine mapping (2018, methodology): “A more sophisticated version of this idea is to use hierarchical testing (Meinshausen, 2008; Yekutieli, 2008; Mandozzi and Bühlmann, 2016; Renaux et al., 2018), which requires specification of a hierarchy on the variables, but avoids an a priori decision on where to draw group boundaries.”
Multiple Testing for Exploratory Research (2011, methodology): “Examples of such procedures are the tree-based testing procedure of Meinshausen (2008), recently improved by Goeman and Solari (2010), the focus level procedure of Goeman and Mansmann (2008), and the gatekeeping method of Edwards and Madsen (2007).”
... and the Cross-Section of Expected Returns (2013, unknown): “See Meinshausen (2008) for a hierarchical approach on variable selection.”
Multiple hypothesis testing in genomics (2014, methodology): “Hierarchical methods are available for both FDR and FWER control [92–96].”