4期

最新文章

• Optimal alpha spending for sequential analysis with binomial data AI翻译
  作者：Silva, Ivair R.; Kulldorff, Martin; Yih, W. Katherine
  作者单位：Universidade Federal de Ouro Preto; Harvard University; Harvard Medical School; Harvard University; Harvard University Medical Affiliates; Brigham & Women's Hospital; Harvard Pilgrim Health Care
  摘要：For sequential analysis hypothesis testing, various alpha spending functions have been proposed. Given a prespecified overall alpha level and power, we derive the optimal alpha spending function that minimizes the expected time to signal for continuous as well as group sequential analysis. If there is also a restriction on the maximum sample size or on the expected sample size, we do the same. Alternatively, for fixed overall alpha, power and expected time to signal, we derive the optimal alph...
• Quasi-Bayes properties of a procedure for sequential learning in mixture models AI翻译
  作者：Fortini, Sandra; Petrone, Sonia
  作者单位：Bocconi University
  摘要：Bayesian methods are often optimal, yet increasing pressure for fast computations, especially with streaming data, brings renewed interest in faster, possibly suboptimal, solutions. The extent to which these algorithms approximate Bayesian solutions is a question of interest, but often unanswered. We propose a methodology to address this question in predictive settings, when the algorithm can be reinterpreted as a probabilistic predictive rule. We specifically develop the proposed methodology ...
• A scalable estimate of the out-of-sample prediction error via approximate leave-one-out cross-validation AI翻译
  作者：Rad, Kamiar Rahnama; Maleki, Arian
  作者单位：City University of New York (CUNY) System; Columbia University
  摘要：The paper considers the problem of out-of-sample risk estimation under the high dimensional settings where standard techniques such asK-fold cross-validation suffer from large biases. Motivated by the low bias of the leave-one-out cross-validation method, we propose a computationally efficient closed form approximate leave-one-out formula ALO for a large class of regularized estimators. Given the regularized estimate, calculating ALO requires a minor computational overhead. With minor assumpti...
• False discovery and its control in low rank estimation AI翻译
  作者：Taeb, Armeen; Shah, Parikshit; Chandrasekaran, Venkat
  作者单位：California Institute of Technology; Yahoo! Inc
  摘要：Models specified by low rank matrices are ubiquitous in contemporary applications. In many of these problem domains, the row-column space structure of a low rank matrix carries information about some underlying phenomenon, and it is of interest in inferential settings to evaluate the extent to which the row-column spaces of an estimated low rank matrix signify discoveries about the phenomenon. However, in contrast with variable selection, we lack a formal framework to assess true or false disc...
• A unified data-adaptive framework for high dimensional change point detection AI翻译
  作者：Liu, Bin; Zhou, Cheng; Zhang, Xinsheng; Liu, Yufeng
  作者单位：Fudan University; Tencent; University of North Carolina; University of North Carolina Chapel Hill
  摘要：In recent years, change point detection for a high dimensional data sequence has become increasingly important in many scientific fields such as biology and finance. The existing literature develops a variety of methods designed for either a specified parameter (e.g. the mean or covariance) or a particular alternative pattern (sparse or dense), but not for both scenarios simultaneously. To overcome this limitation, we provide a general framework for developing tests that are suitable for a lar...
• Graphical models for extremes AI翻译
  作者：Engelke, Sebastian; Hitz, Adrien S.
  作者单位：University of Geneva; University of Oxford
  摘要：Conditional independence, graphical models and sparsity are key notions for parsimonious statistical models and for understanding the structural relationships in the data. The theory of multivariate and spatial extremes describes the risk of rare events through asymptotically justified limit models such as max-stable and multivariate Pareto distributions. Statistical modelling in this field has been limited to moderate dimensions so far, partly owing to complicated likelihoods and a lack of un...
• Adaptive designs for optimal observed Fisher information AI翻译
  作者：Lane, Adam
  作者单位：Cincinnati Children's Hospital Medical Center
  摘要：Expected Fisher information can be founda prioriand as a result its inverse is the primary variance approximation used in the design of experiments. This is in contrast with the common claim that the inverse of the observed Fisher information is a better approximation of the variance of the maximum likelihood estimator. Observed Fisher information cannot be knowna priori; however, if an experiment is conducted sequentially, in a series of runs, the observed Fisher information from previous run...
• Superconsistent estimation of points of impact in non-parametric regression with functional predictors AI翻译
  作者：Poss, Dominik; Liebl, Dominik; Kneip, Alois; Eisenbarth, Hedwig; Wager, Tor D.; Barrett, Lisa Feldman
  作者单位：University of Bonn; Victoria University Wellington; Dartmouth College; Northeastern University; Harvard University; Harvard University Medical Affiliates; Massachusetts General Hospital; Harvard University; Harvard Medical School; Harvard University; Harvard University Medical Affiliates; Massachusetts General Hospital
  摘要：Predicting scalar outcomes by using functional predictors is a classical problem in functional data analysis. In many applications, however, only specific locations or time points of the functional predictors have an influence on the outcome. Such 'points of impact' are typically unknown and must be estimated in addition to estimating the usual model components. We show that our points-of-impact estimator enjoys a superconsistent rate of convergence and does not require knowledge or pre-estima...
• Visualizing the effects of predictor variables in black box supervised learning models AI翻译
  作者：Apley, Daniel W.; Zhu, Jingyu
  作者单位：Northwestern University
  摘要：In many supervised learning applications, understanding and visualizing the effects of the predictor variables on the predicted response is of paramount importance. A shortcoming of black box supervised learning models (e.g. complex trees, neural networks, boosted trees, random forests, nearest neighbours, local kernel-weighted methods and support vector regression) in this regard is their lack of interpretability or transparency. Partial dependence plots, which are the most popular approach f...