Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313643653> ?p ?o ?g. }
- W4313643653 endingPage "107689" @default.
- W4313643653 startingPage "107689" @default.
- W4313643653 abstract "Hypothesis testing of variable importance measures (VIMPs) is still the subject of ongoing research. This particularly applies to random forests (RF), for which VIMPs are a popular feature. Among recent developments, heuristic approaches to parametric testing have been proposed whose distributional assumptions are based on empirical evidence. Other formal tests under regularity conditions were derived analytically. But these approaches can be computationally expensive or even practically infeasible. This problem also occurs with non-parametric permutation tests, which are, however, distribution-free and can generically be applied to any kind of prediction model and VIMP. Embracing this advantage, it is proposed to use sequential permutation tests and sequential p-value estimation to reduce the computational costs associated with conventional permutation tests. These costs can be particularly high in case of complex prediction models. Therefore, RF's popular and widely used permutation VIMP (pVIMP) serves as a practical and relevant application example. The results of simulation studies confirm the theoretical properties of the sequential tests, that is, the type-I error probability is controlled at a nominal level and a high power is maintained with considerably fewer permutations needed compared to conventional permutation testing. The numerical stability of the methods is investigated in two additional application studies. In summary, theoretically sound sequential permutation testing of VIMP is possible at greatly reduced computational costs. Recommendations for application are given. A respective implementation for RF's pVIMP is provided through the accompanying R package rfvimptest." @default.
- W4313643653 created "2023-01-07" @default.
- W4313643653 creator A5042014267 @default.
- W4313643653 creator A5048970390 @default.
- W4313643653 creator A5086861344 @default.
- W4313643653 date "2023-05-01" @default.
- W4313643653 modified "2023-10-02" @default.
- W4313643653 title "Efficient permutation testing of variable importance measures by the example of random forests" @default.
- W4313643653 cites W1602160603 @default.
- W4313643653 cites W1875061881 @default.
- W4313643653 cites W1969255591 @default.
- W4313643653 cites W1973886547 @default.
- W4313643653 cites W1978263273 @default.
- W4313643653 cites W1979279626 @default.
- W4313643653 cites W1984941527 @default.
- W4313643653 cites W1985372952 @default.
- W4313643653 cites W1999795754 @default.
- W4313643653 cites W2017665047 @default.
- W4313643653 cites W2023071006 @default.
- W4313643653 cites W2046793639 @default.
- W4313643653 cites W2070230130 @default.
- W4313643653 cites W2105398277 @default.
- W4313643653 cites W2110878112 @default.
- W4313643653 cites W2123998733 @default.
- W4313643653 cites W2143481518 @default.
- W4313643653 cites W2166661898 @default.
- W4313643653 cites W2181164912 @default.
- W4313643653 cites W2261059368 @default.
- W4313643653 cites W2322754609 @default.
- W4313643653 cites W2515235622 @default.
- W4313643653 cites W2608146202 @default.
- W4313643653 cites W2766438525 @default.
- W4313643653 cites W2802643674 @default.
- W4313643653 cites W2806446597 @default.
- W4313643653 cites W2905206498 @default.
- W4313643653 cites W2911964244 @default.
- W4313643653 cites W2944954104 @default.
- W4313643653 cites W2973941913 @default.
- W4313643653 cites W3009833198 @default.
- W4313643653 cites W3043698154 @default.
- W4313643653 cites W3102027041 @default.
- W4313643653 cites W3124574412 @default.
- W4313643653 cites W3200824799 @default.
- W4313643653 cites W3209434414 @default.
- W4313643653 cites W3215376663 @default.
- W4313643653 cites W4212883601 @default.
- W4313643653 cites W4230704549 @default.
- W4313643653 cites W4231210484 @default.
- W4313643653 cites W4301594934 @default.
- W4313643653 doi "https://doi.org/10.1016/j.csda.2022.107689" @default.
- W4313643653 hasPublicationYear "2023" @default.
- W4313643653 type Work @default.
- W4313643653 citedByCount "4" @default.
- W4313643653 countsByYear W43136436532023 @default.
- W4313643653 crossrefType "journal-article" @default.
- W4313643653 hasAuthorship W4313643653A5042014267 @default.
- W4313643653 hasAuthorship W4313643653A5048970390 @default.
- W4313643653 hasAuthorship W4313643653A5086861344 @default.
- W4313643653 hasBestOaLocation W43136436531 @default.
- W4313643653 hasConcept C105795698 @default.
- W4313643653 hasConcept C11413529 @default.
- W4313643653 hasConcept C117251300 @default.
- W4313643653 hasConcept C118615104 @default.
- W4313643653 hasConcept C121332964 @default.
- W4313643653 hasConcept C126255220 @default.
- W4313643653 hasConcept C128622974 @default.
- W4313643653 hasConcept C134306372 @default.
- W4313643653 hasConcept C150921843 @default.
- W4313643653 hasConcept C173801870 @default.
- W4313643653 hasConcept C182365436 @default.
- W4313643653 hasConcept C200985842 @default.
- W4313643653 hasConcept C21308566 @default.
- W4313643653 hasConcept C24890656 @default.
- W4313643653 hasConcept C33923547 @default.
- W4313643653 hasConcept C40696583 @default.
- W4313643653 hasConcept C41008148 @default.
- W4313643653 hasConcept C87007009 @default.
- W4313643653 hasConceptScore W4313643653C105795698 @default.
- W4313643653 hasConceptScore W4313643653C11413529 @default.
- W4313643653 hasConceptScore W4313643653C117251300 @default.
- W4313643653 hasConceptScore W4313643653C118615104 @default.
- W4313643653 hasConceptScore W4313643653C121332964 @default.
- W4313643653 hasConceptScore W4313643653C126255220 @default.
- W4313643653 hasConceptScore W4313643653C128622974 @default.
- W4313643653 hasConceptScore W4313643653C134306372 @default.
- W4313643653 hasConceptScore W4313643653C150921843 @default.
- W4313643653 hasConceptScore W4313643653C173801870 @default.
- W4313643653 hasConceptScore W4313643653C182365436 @default.
- W4313643653 hasConceptScore W4313643653C200985842 @default.
- W4313643653 hasConceptScore W4313643653C21308566 @default.
- W4313643653 hasConceptScore W4313643653C24890656 @default.
- W4313643653 hasConceptScore W4313643653C33923547 @default.
- W4313643653 hasConceptScore W4313643653C40696583 @default.
- W4313643653 hasConceptScore W4313643653C41008148 @default.
- W4313643653 hasConceptScore W4313643653C87007009 @default.
- W4313643653 hasLocation W43136436531 @default.
- W4313643653 hasLocation W43136436532 @default.
- W4313643653 hasOpenAccess W4313643653 @default.