Prediction intervals - a review

This review covers some known results on prediction intervals for univariate distributions. Results for parametric continuous and discrete distributions as well as those based on distribution-free methods are included. Prediction intervals based on Bayesian and sequential methods are not covered. Methods of construction of prediction intervals and other related problems are discussed.     

Mem squared error estimation in finite populations under nonlinear models

Four estimators of the prediction mean squared error (MSB) of an estimated finite population total for a zero-one characteristic are examined. The characteristic associated with each population unit is modeled as the realization of a Bernoulli random variable whose expected value is a nonlinear function of a parameter vector and a set of known auxiliary variables. To compare the estimators, a simulation study is conducted using a population of hospitals. The MSB estimator Implied by the form of the assumed model underestimates the mean squared error in each of the cases studied and produces confidence lntervals with less than the nominal coverage probabilities. Of the three alternative MSE estimators presented, a linear approximation to the jackknife produces the best results and improves upon the model-specific estimator.     

Bayesian prediction with a random sample size for the burr lifetime distribution

This paper is concerned with the problem of deriving Bayesian prediction bounds for the Burr distribution when the sample size is a random variable. Prediction bounds for both the future observations (the case of two-sample prediction) and the remaining observations in the same sample (the case of one-sample prediction) will be derived. The analysis will depend mainly on assuming that the size of the sample is a random variable having the Poisson distribution. Finally, numerical examples are given to illustrate the results.     

Factors effecting the model performance measures area under the ROC curve,net reclassification improvement and integrated discrimination improvement

ABSTRACT

The aim of this study is to investigate the impact of correlation structure, prevalence and effect size on the risk prediction model by using the change in the area under the receiver operating characteristic curve (ΔAUC), net reclassification improvement (NRI), and integrated discrimination improvement (IDI). In simulation study, the dataset is generated under different correlation structures, prevalences and effect sizes. We verify the simulation results with the real-data application. In conclusion, the correlation structure between the variables should be taken into account while composing a multivariable model. Negative correlation structure between independent variables is more beneficial while constructing a model.     

A Multivariate Quantile Predictor

ABSTRACT

We introduce a nonparametric quantile predictor for multivariate time series via generalizing the well-known univariate conditional quantile into a multivariate setting for dependent data. Applying the multivariate predictor to predicting tail conditional quantiles from foreign exchange daily returns, it is observed that the accuracy of extreme tail quantile predictions can be greatly improved by incorporating interdependence between the returns in a bivariate framework. As a special application of the multivariate quantile predictor, we also introduce a so-called joint-horizon quantile predictor that is used to produce multi-step quantile predictions in one-go from univariate time series realizations. A simulation example is discussed to illustrate the relevance of the joint-horizon approach.     

Assessing One-Step-Ahead Prediction Error Based on the Median for First-Order Autoregressive Models in the Presence Of Outliers

The prediction of the one-step-ahead observation of the first-order autoregressive process in the presence of outliers is considered. The mean square of the prediction error is obtained based on the median estimator of the model parameter for a stationary process. Monte Carlo simulation methods are employed to investigate the performance of the proposed estimator as well as the conventional ordinary least squares estimators proposed by Zhang and Shaman (1995 Zhang , P. , Shaman , P. ( 1995 ). Assessing prediction error in autoregressive models . Trans. Amer. Mathemat. Soc. 347 : 627 – 637 .[Crossref], [Web of Science ®] , [Google Scholar]) and Kabaila and He (1999 Kabaila , P. , He , Z. ( 1999 ). On assessing prediction error in autoregressive models . J. Time Ser. Anal. 20 : 663 – 670 .[Crossref] , [Google Scholar]) for a process without outliers. The results show that the proposed method outperforms the conventional method. These conclusions are substantiated with results from actual datasets.     

Nonparametric Prediction Intervals Based on Ranked Set Samples

In this article, we develop nonparametric prediction intervals based on generalized ranked set samples using conditional as well as unconditional approaches. The predictions are developed for order statistics from a future sample as well as for order statistics from a future balanced ranked set sample. The effects of ranking errors on the coverage probabilities of these prediction intervals are also examined.     

Spatial Regression Models Using Inter-Region Distances in a Non-Random Context

This article considers spatial data z( s ₁), z( s ₂),…, z( s _n ) collected at n locations, with the objective of predicting z( s ₀) at another location. The usual method of analysis for this problem is kriging, but here we introduce a new signal-plus-noise model whose essential feature is the identification of hot spots. The signal decays in relation to distance from hot spots. We show that hot spots can be located with high accuracy and that the decay parameter can be estimated accurately. This new model compares well to kriging in simulations.     

Prediction in Multilevel Logistic Regression

The purpose of this article is to present a new method to predict the response variable of an observation in a new cluster for a multilevel logistic regression. The central idea is based on the empirical best estimator for the random effect. Two estimation methods for multilevel model are compared: penalized quasi-likelihood and Gauss–Hermite quadrature. The performance measures for the prediction of the probability for a new cluster observation of the multilevel logistic model in comparison with the usual logistic model are examined through simulations and an application.     

不报、谎报安全事故罪的司法适用问题研究

2006年通过的《中华人民共和国刑法修正案（六）》新增了不报、谎报安全事故罪。要认定本罪,应注意理解＂不报或者谎报＂、＂贻误事故抢救＂及＂情节严重＂三个要素;本罪与其他安全事故犯罪发生关联适用时,应数罪并罚;本罪与滥用职权罪或玩忽职守罪发生关联适用时,应对不报、谎报安全事故情况的国家机关工作人员以滥用职权罪论处。