A time series bootstrap procedure for interpolation intervals

A sieve bootstrap procedure for constructing interpolation intervals for a general class of linear processes is proposed. This sieve bootstrap provides consistent estimators of the conditional distribution of the missing values, given the observed data. A Monte Carlo experiment is used to show the finite sample properties of the sieve bootstrap and finally, the performance of the proposed method is illustrated with a real data example.     

On the binomial confidence interval and probabilistic robust control

The Clopper-Pearson confidence interval has ever been documented as an exact approach in some statistics literature. More recently, such approach of interval estimation has been introduced to probabilistic control theory and has been referred as non-conservative in control community. In this note, we clarify the fact that the so-called exact approach is actually conservative. In particular, we derive analytic results demonstrating the extent of conservatism in the context of probabilistic robustness analysis. This investigation encourages seeking better methods of confidence interval construction for robust control purpose.     

评定飞航导弹单发命中概率的方法研究

通过对抽样检验方法的研究,确立了区间估计的下限公式,用来在已知样本值的情况下确定导弹单发命中概率出现在该区间内的置信水平和精确度。     

How Meta-Analysis Increases Statistical Power.

One of the most frequently cited reasons for conducting a meta-analysis is the increase in statistical power that it affords a reviewer. This article demonstrates that fixed-effects meta-analysis increases statistical power by reducing the standard error of the weighted average effect size (T?.) and, in so doing, shrinks the confidence interval around T?.. Small confidence intervals make it more likely for reviewers to detect nonzero population effects, thereby increasing statistical power. Smaller confidence intervals also represent increased precision of the estimated population effect size. Computational examples are provided for 3 effect-size indices: d (standardized mean difference), Pearson's r, and odds ratios. Random-effects meta-analyses also may show increased statistical power and a smaller standard error of the weighted average effect size. However, the authors demonstrate that increasing the number of studies in a random-effects meta-analysis does not always increase statistical power. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Distinctive Features of Minimization of a Risk Functional in Mass Data Sets

A statistical learning model is considered within the framework of the theory of uniform convergence of frequencies of errors in the case where the convergence is violated as a result of increasing the informativeness of training examples. Drawbacks of nonconstructive refinements of Vapnik-Chervonenkis estimates based on an assumption on the distribution law of violations are shown. A new approach to obtaining constructive estimates for mass data sets is proposed.     

Measuring the errors of form of industrial surfaces: prediction and optimization

To acquire maximum information on the geometrical errors of industrially made surfaces at a minimum cost, a method for estimating conditional probabilities of a random signal (Bayesian prediction) is applied to three-dimensional metrology. First, a surface is interpolated between data acquired on a coordinate measuring machine (CMM). Then, for a given probability, limit surfaces are computed that bind a region of space containing the known data and the most probable interpolation of the missing data of the surface. These bounds can be treated as the surface; their points can be considered as if they were actual CMM data when fitting a tolerance zone or a datum feature to the data. For Bayesian prediction, the basic hypotheses on the signal are stationarity, ergodicity, and gaussian density. Deviations from these hypotheses and their consequences on the prediction are taken into account and corrections are proposed.     

Parameter identification for nonlinear systems: Guaranteed confidence regions through LSCR

In this paper we consider the problem of constructing confidence regions for the parameters of nonlinear dynamical systems. The proposed method uses higher order statistics and extends the LSCR (leave-out sign-dominant correlation regions) algorithm for linear systems introduced in Campi and Weyer [2005, Guaranteed non-asymptotic confidence regions in system identification. Automatica 41(10), 1751-1764. Extended version available at 〈http://www.ing.unibs.it/∼campi〉]. The confidence regions contain the true parameter value with a guaranteed probability for any finite number of data points. Moreover, the confidence regions shrink around the true parameter value as the number of data points increases. The usefulness of the proposed approach is illustrated on some simple examples.