R
Ronald H. Randles
Researcher at University of Iowa
Publications - 12
Citations - 500
Ronald H. Randles is an academic researcher from University of Iowa. The author has contributed to research in topics: Optimal discriminant analysis & Kernel Fisher discriminant analysis. The author has an hindex of 10, co-authored 12 publications receiving 494 citations.
Papers
More filters
Journal ArticleDOI
A Two-Sample Adaptive Distribution-Free Test
TL;DR: In this article, an adaptive distribution-free test is proposed for the two-sample location problem, where the data are used to assess the tailweight and skewness of the underlying distributions, leading to the selection and then application, with the same data, of one of several common rank tests for shift, such as the Mann-Whitney-Wilcoxon test.
Journal ArticleDOI
Generalized Linear and Quadratic Discriminant Functions Using Robust Estimates
TL;DR: In this paper, a generalization of Fisher's procedure for finding a linear discriminant function was proposed, which places less weight on those observations that are far from the overlapping regions of the two populations.
Journal ArticleDOI
Adaptive distribution-free tests
Ronald H. Randles,Robert V. Hogg +1 more
TL;DR: In this paper, some adaptive distribution-free tests about location parameters of symmetric distributions are processed for the one-and two sample cases, and preliminary inferences in these adaptive schemes are to classify the underlying distributions as having light, medium, or heave rails with a statistic that is assentialy the range of the sample divided by the mean deviation from the sample median.
Journal ArticleDOI
Distribution-Free Partial Discriminant Analysis
TL;DR: In this paper, a distribution-free rank procedure was proposed for partial discrimination problems involving two populations, which can be applied with virtually any discriminant function and is shown to be suitable for any classifier.
Journal ArticleDOI
Discriminant Analysis Based on Ranks
TL;DR: In this paper, a model-free rank procedure is proposed for two-population discrimination problem that enables the practitioner to better control the balance between the two probabilities of misclassification.