ABSTRACTS

This article provides methods for constructing simultaneous prediction intervals to contain the means of the dependent variable in a regression model for each of k future samples at k sets of values of the independent variables, some or all of which may be different. The methods are compared with previously proposed approximate procedures. The construction of simultaneous confidence intervals to contain the true regression at all of k vectors of independent variables is also presented.     

并行工程环境下的产品质量监视系统

提出了一种在并行工程环境下用于小批量生产的质量反馈两层模型，能够准确，客观、在线动态地评价现行加工工序的状况和进行相应的加工过程质量仿真。     

On the use of time series techniques in shape recognition applications

In this paper, a method for classifying objects based on the use of autoregressive model parameters which are obtained from a time series representation of shape boundaries in digital images of objects is presented. This technique is insensitive to size and is rotation invariant. The objects chosen are four types of aircraft from a digital photograph. Recognition accuracies of more than 90% were obtained for all the pattern classes. All pattern recognition problems involve two random variables, the pattern vector and the class to which it belongs. The interdependence of the two variables is given by the conditional probability density function. The degree of dependence between the pattern vector and the particular class is measured by the “distance”. A simple Bhattacharya distance classifier was used for the purpose.     

Uniform Confidence Bands for a Quadratic Model

In this paper uniform confidence bands are developed for a quadratic regression model. Tables are supplied to determine a uniform confidence band for the quadratic model in which it is assumed the sum of the cubed deviations of the independent variable about its mean is equal to zero. The uniform confidence band is then compared with the classical competitor, the Scheffit confidence band. A table summarizing the results of this comparison is presented. Finally, an example is examined.     

Estimation of the Fraction Defective in Curtailed Sampling Plans by Attributes

In this paper we have obtained the maximum likelihood estimate of the fraction defective (p) under three different sampling plans. Plan 1 considers complete inspection of the sample, Plan 2 considers curtailing of the inspection at the rejection stage and Plan 3 considers curtailing of the inspection at both the acceptance and rejection stages.     

A Prediction Interval on the Difference Between Two Future Sample Means and its Application to a Claim of Product Superiority

This paper provides procedures for constructing a prediction interval to contain the difference between single future observations from each of two previously sampled normal populations, or, more generally, to contain the difference between two future sample means.

An application is the situation where a producer has sampled the performance of both his product and that of a competitor, and wishes to obtain intervals to contain: i. The difference in future performance to be observed by a consumer who will purchase one item of each product.

ii. The difference in future mean performance to be obtained by a regulating agency which will sample a number of items from each of the two products.

This paper describes the problem, gives solutions for both the situation where the two populations have the same standard deviation and where they do not, provides a numerical example, and references related work. The mathematical justification is given in an Appendix.     

On statistical inference in time series analysis of the evolution of road safety

Data collected for building a road safety observatory usually include observations made sequentially through time. Examples of such data, called time series data, include annual (or monthly) number of road traffic accidents, traffic fatalities or vehicle kilometers driven in a country, as well as the corresponding values of safety performance indicators (e.g., data on speeding, seat belt use, alcohol use, etc.). Some commonly used statistical techniques imply assumptions that are often violated by the special properties of time series data, namely serial dependency among disturbances associated with the observations. The first objective of this paper is to demonstrate the impact of such violations to the applicability of standard methods of statistical inference, which leads to an under or overestimation of the standard error and consequently may produce erroneous inferences. Moreover, having established the adverse consequences of ignoring serial dependency issues, the paper aims to describe rigorous statistical techniques used to overcome them. In particular, appropriate time series analysis techniques of varying complexity are employed to describe the development over time, relating the accident-occurrences to explanatory factors such as exposure measures or safety performance indicators, and forecasting the development into the near future. Traditional regression models (whether they are linear, generalized linear or nonlinear) are shown not to naturally capture the inherent dependencies in time series data. Dedicated time series analysis techniques, such as the ARMA-type and DRAG approaches are discussed next, followed by structural time series models, which are a subclass of state space methods. The paper concludes with general recommendations and practice guidelines for the use of time series models in road safety research.     

Prediction of driving ability: Are we building valid models?

The prediction of on-road driving ability using off-road measures is a key aim in driving research. The primary goal in most classification models is to determine a small number of off-road variables that predict driving ability with high accuracy. Unfortunately, classification models are often over-fitted to the study sample, leading to inflation of predictive accuracy, poor generalization to the relevant population and, thus, poor validity. Many driving studies do not report sufficient details to determine the risk of model over-fitting and few report any validation technique, which is critical to test the generalizability of a model. After reviewing the literature, we generated a model using a moderately large sample size (n = 279) employing best practice techniques in the context of regression modelling. By then randomly selecting progressively smaller sample sizes we show that a low ratio of participants to independent variables can result in over-fitted models and spurious conclusions regarding model accuracy. We conclude that more stable models can be constructed by following a few guidelines.     

超精密车削表面粗糙度预测模型的建立

介绍了一种利用回归分析法来建立单点金刚石刀具超精车削表面粗糙度预测模型的新方法,并通过建立的粗糙度预测模型,研究了铝合金超精密车削过程中切削速度,进给量和切削深度等参数对表面粗糙度的影响。通过实验分析表明：二次预测方程比一次预测方程更有效,而且适用范围比一次模型大。利用优化设计中的约束变尺度法对所建立的表面粗糙度预测方程进行了优化,可以实现对切削参数的优选,从而达到加工前在特定的条件下预测和控制表     

An alternative derivation of the distribution of the individual bioequivalence metric

A method is proposed for estimating the probability density function (PDF) of the individual bioequivalence metric. We show that under the usual assumption of independence of the various random variables, the computations are quite feasible. A program to evaluate the integrals needed for the computations has been written. Experimentation with this program shows that the pass/fail decision based on the computed PDF compares favorably with that based on the method suggested in a recent FDA Guidance (Jan. 2001).     

Inference on the Gamma Distribution

This study develops inferential procedures for a gamma distribution. Based on the Cornish–Fisher expansion and pivoting the cumulative distribution function, an approximate confidence interval for the gamma shape parameter is derived. The generalized confidence intervals for the rate parameter and other quantities such as mean are explored. The proposed generalized inferential procedures are extended to construct prediction limits for a single future measurement and for at least p of m measurements at each of r locations. The performance of the proposed procedures is evaluated using Monte Carlo simulation. The simulation results show that the proposed procedures are very satisfactory. Finally, three real examples are used to illustrate the proposed procedures. Supplementary materials for this article are available online.     

An Alternative Derivation of the Distribution of the Individual Bioequivalence Metric

Abstract

A method is proposed for estimating the probability density function (PDF) of the individual bioequivalence metric. We show that under the usual assumption of independence of the various random variables, the computations are quite feasible. A program to evaluate the integrals needed for the computations has been written. Experimentation with this program shows that the pass/fail decision based on the computed PDF compares favorably with that based on the method suggested in a recent FDA Guidance (Jan. 2001).     

Neural Networks in Applied Statistics

Artificial neural networks are computer algorithms or computer programs derived in part from attempts to model the activity of nerve cells. They have been applied to pattern recognition, classification, and optimization problems in the physical and chemical sciences, as well as in other fields. We introduce the principles of the multilayer feedforward network that is among the most commonly used neural networks in practical problems. The relevance of neural network models for the applied statistician is considered using a time series prediction problem as an example. The multilayer feedforward neural network uses a nonlinear function of the predictors to obtain predictions for future time series values. We illustrate the considerations involved in specifying a neural network model and evaluate the accuracy of neural network models relative to the accuracy obtained using other computer-intensive, nonmodel-based techniques.     

Uncertainty analysis of reliability predictions for brittle fracture

Reliability analysis of ceramic components under stationary or transient loading is generally performed on the basis of a Finite Element stress analysis from which the failure probability according to the multi-axial Weakest Link theory is calculated with the help of a suitable post-processing routine. We use the STAU post-processing routine and the general purpose Finite Element code ABAQUS. Due to scatter in the material parameters, the resulting failure probability is also prone to statistical uncertainties. We present a method of assessing this scatter using so-called resampling simulation methods. The analysis leads to confidence intervals for the failure probability which is a novel and important result especially for the purpose of design sensitivity considerations and the assessment of pooling procedures. In a simple example using a four-point bend specimen, the effect of pooling (i.e. grouping of results from different experiments by suitable scaling procedures) on the numerical result and on the scatter of failure probability is demonstrated. Here, pooling is done using results of inert strength measurements at various temperatures and scaling to room temperature values. A technologically more relevant example deals with a ceramic component in a model clutch under thermo-mechanical frictional loading. As a first step, the local risk of rupture is calculated which leads to the identification of the most critical regions of the component. As a second step, resampling confidence intervals for the failure probability are determined. As resampling data base, we use inert strength values at different temperatures as well as material data for sub-critical crack propagation.     

Sparse partial least-squares regression and its applications to high-throughput data analysis

The partial least-squares (PLS) method is designed for prediction problems where the number of predictors is larger than the number of training samples. PLS is based on latent components that are linear combinations of all of the original predictors, so it automatically employs all predictors regardless of their relevance. This will potentially compromise its performance, but it will also make it difficult to interpret the result. In this paper, we propose a new formulation of the sparse PLS (SPLS) procedure to allow both sparse variable selection and dimension reduction. We use the standard L₁-penalty and the unbounded penalty of [1]. We develop a computing algorithm for SPLS by modifying the nonlinear iterative partial least-squares (NIPALS) algorithm, and illustrate the method with an analysis of a cancer dataset. Through the numerical studies we find that our SPLS method generally performs better than the standard PLS and other existing methods in variable selection and prediction.     

Forecasting Trending Time Series with Relative Growth Rate Models

Many annual time series in socioeconomic systems are steadily increasing functions of time. This paper deals with an empirical approach to analyzing and projecting such trending time series from models of relative growth rates or percent changes. A class of relative growth rate models is defined which includes the linear. exponential, modified exponential and logistic growth curves as special eases. Parameters are estimated for the most part by linear regression techniques since the, relative growth rates for this class of models are linear in the parameters. The ternd curve for each model is obtained by integration and approximate-confidence limits can be obtained for the forecasts of future series values.     

时变参数识别方法研究

本文重点研究了现有各种时变参数识别方法的内在联系及各自特点。在引起统一的模型结构以后，利用随机逼近原理，提出了各种在线识别算法的一般递推形式。然后基于变物理参数微分方程模型，推导出了不同辨识算法各自的递推形式。用六种方法对参数缓变，快变及突变三种系统作了计算机仿真识别，以比较各种的特点。研究结果对时变参数识别具有理论和应用价值。     

Adapting ISA system warnings to enhance user acceptance

Inappropriate speed is a major cause of traffic accidents. Different measures have been considered to control traffic speed, and intelligent speed adaptation (ISA) systems are one of the alternatives. These systems know the speed limits and try to improve compliance with them. This paper deals with an informative ISA system that provides the driver with an advance warning before reaching a road section with singular characteristics that require a lower safe speed than the current speed. In spite of the extensive tests performed using ISA systems, few works show how warnings can be adapted to the driver. This paper describes a method to adapt warning parameters (safe speed on curves, zone of influence of a singular stretch, deceleration process and reaction time) to normal driving behavior. The method is based on a set of tests with and without the ISA system. This adjustment, as well as the analysis of driver acceptance before and after the adaptation and changes in driver behavior (changes in speed and path) resulting from the tested ISA regarding a driver's normal driving style, is shown in this paper. The main conclusion is that acceptance by drivers increased significantly after redefining the warning parameters, but the effect of speed homogenization was not reduced.