Shorter nonparametric prediction intervals for an order statistic from a future sample

Standard nonparametric prediction intervals for a future order statistic are obtained by taking the interval between two order statistics of the initial sample. We obtain improved prediction intervals by taking the shortest of two or more standard intervals.     

Bayesian forecasting with the Holt–Winters model

Exponential smoothing methods are widely used as forecasting techniques in inventory systems and business planning, where reliable prediction intervals are also required for a large number of series. This paper describes a Bayesian forecasting approach based on the Holt–Winters model, which allows obtaining accurate prediction intervals. We show how to build them incorporating the uncertainty due to the smoothing unknowns using a linear heteroscedastic model. That linear formulation simplifies obtaining the posterior distribution on the unknowns; a random sample from such posterior, which is not analytical, is provided using an acceptance sampling procedure and a Monte Carlo approach gives the predictive distributions. On the basis of this scheme, point-wise forecasts and prediction intervals are obtained. The accuracy of the proposed Bayesian forecasting approach for building prediction intervals is tested using the 3003 time series from the M3-competition.     

Nonparametric prediction intervals for future record intervals based on order statistics

In this article, outer and inner prediction intervals for future record intervals as well as record spacings are derived based on observed order statistics from the same parent distribution. These intervals are exact and are distribution-free in that they do not depend on the sampling distribution. Three different cases are considered and in each case an exact explicit expression is obtained for the prediction coefficient. Finally, we compare the obtained results with similar intervals based on records, and also present a numerical example in order to illustrate the derived results.     

Optimal plan for Wiener constant-stress accelerated degradation model

This paper explores inferential procedures for the Wiener constant-stress accelerated degradation model under degradation mechanism invariance. The exact confidence intervals are obtained for the parameters of the proposed accelerated degradation model. The generalized confidence intervals are also proposed for the reliability function and pth quantile of the lifetime at the normal operating stress level. In addition, the prediction intervals are developed for the degradation characteristic, lifetime and remaining useful life of the product at the normal operating stress level. The performance of the proposed generalized confidence intervals and the prediction intervals is assessed by the Monte Carlo simulation. Furthermore, a new optimum criterion is proposed based on minimizing the mean of the upper prediction limit for the degradation characteristic at the design stress level. The exact optimum plan is also derived for the Wiener accelerated degradation model according to the proposed optimal criterion. The proposed interval procedures and optimum plan are the free of the equal testing interval assumption. Finally, two examples are provided to illustrate the proposed interval procedures and exact optimum plan. Specifically, based on the degradation data of LEDs, some interval estimates of quantities related to reliability indicators are obtained. For the degradation data of carbon-film resistors, the optimal allocation of test units is derived in terms of the proposed optimal criterion.     

Weibull过程将来第k次失效时间的Bayes预测区间

具Weibull强度函数的非齐次Poisson过程经常被用来分析可修系统的失效模式.基于极大似然估计,Engelhardt ＆ Bain(1978)导出了Weibull过程将来第k次失效时间的经典预测区间.在本文中,我们用无信息联合验前分布,根据Weibull过程的前n次失效时间,给出了建立将来第k次失效时间的Bayes预测区间的方法,并说明了如何应用这些方法。     

Prediction and calibration in generalized linear models

This paper concerns prediction and calibration in generalized linear models. A new predictive procedure, giving improved prediction intervals, is briefly reviewed and further theoretical results, useful for calculations, are presented. Indeed, the calibration problem is faced within the classical approach and a suitable solution is obtained by inverting the associated improved prediction procedure. This calibration technique gives accurate confidence regions and it constitutes a substantial improvement over both the estimative solution and the naive solution, which involves, even for non-linear and non-normal models, the results available for the linear Gaussian case. Finally, some useful explicit formulae for the construction of prediction and calibration intervals are presented, with regard to generalized linear models with alternative error terms and link functions. This research was partially supported by a grant from Ministero dell’Instruzione, dell’Università e della Ricerca, Italy.     

The probability distribution of tolerance intervals

Tolerance intervals seem to be infrequently used outside engineering. This paper investigates tolerance intervals and relates them to confidence intervals and prediction intervals. It is believed that the introduction of these topics in introductory and intermediate statistics courses would be beneficial to the student.

Tolerance intervals are constructed using observations from the normal and the truncated Cauchy distributions. Both the expected value of the coverage and the probability distribution of the coverage is investigated. Tolerance intervals are also defined for sample data. These are compared numerically to tolerance intervals for populations.     

Time-simultaneous prediction bands: A new look at the uncertainty involved in forecasting mortality

Conventionally, isolated (point-wise) prediction intervals are used to quantify the uncertainty in future mortality rates and other demographic quantities such as life expectancy. A pointwise interval reflects uncertainty in a variable at a single time point, but it does not account for any dynamic property of the time-series. As a result, in situations when the path or trajectory of future mortality rates is important, a band of pointwise intervals might lead to an invalid inference. To improve the communication of uncertainty, a simultaneous prediction band can be used. The primary objective of this paper is to demonstrate how simultaneous prediction bands can be created for prevalent stochastic models, including the Cairns-Blake-Dowd and Lee-Carter models. The illustrations in this paper are based on mortality data from the general population of England and Wales.     

Prediction interval for disease mapping using hierarchical likelihood

In disease mapping, the Bayesian approach is widely used for forming the prediction interval of relative risks. In this paper we propose a hierarchical-likelihood interval for disease mapping, which accounts for the inflation of standard error estimates caused by uncertainty in the estimation of the fixed parameters. Comparison is made with the Bayesian prediction intervals derived from penalized quasi-likelihood and fully Bayesian methods. Through simulation studies, we show that prediction intervals for random effects using hierarchical likelihood maintains the required level.     

Do forecasts expressed as prediction intervals improve production planning decisions?

A number of studies have shown that providing point forecasts to decision makers can lead to improved production planning decisions. However, point forecasts do not convey information about the level of uncertainty that is associated with forecasts. In theory, the provision of prediction intervals, in addition to point forecasts, should therefore lead to further enhancements in decision quality. To test whether this is the case in practice, participants in an experiment were asked to decide on the production levels that were needed to meet the following week’s demand for a series of products. Either underproduction cost twice as much per unit as overproduction or vice versa. The participants were supplied with either a point forecast, a 50% prediction interval, or a 95% prediction interval for the following week’s demand. The prediction intervals did not improve the quality of the decisions and also reduced the propensity of the decision makers to respond appropriately to the asymmetry in the loss function. A simple heuristic is suggested to allow people to make more effective use of prediction intervals. It is found that applying this heuristic to 85% prediction intervals would lead to nearly optimal decisions.     

Nonparametric prediction intervals for progressive Type-II censored order statistics based on k-records

In this paper, we consider the prediction problem in two-sample case and study the non-parametric predicting future progressively Type-II censored order statistics based on observed $k$ -records from the same distribution. Also, prediction intervals for progressively Type-II censored spacings are obtained based on $k$ -record spacings. It is shown that the coverage probabilities of these intervals are exact and do not depend on the underlying distribution. Moreover, optimal prediction intervals are derived for each case. Finally, for illustrating the proposed procedure, we consider a real data set and numerical computations are given. The results of Ahmadi and Balakrishnan (Statistics 44:417–430, 2010) can be achieved as special cases of our results.     

Large sample prediction intervals for future geometric mean a comparative study

This paper provides simulation comparisons among the performance of 11 possible prediction intervals for the geometric.mean of a Pareto distribution with parameters (αB, ). Six different procedures were used to obtain these intervals , namely; true inter -val , pivotal interval , maximum likelihood estimation interval, centrallimit teorem interval, variance stabilizing interval and a mixture of the above intervals . Some of these intervals are valid if the observed sample size m,are large , others are valid if both, n and the future sample size m, are large. Some of these intervals require a knowledge of α or B, while others do not. The simulation validation and efficiency study shows that intervals depending on the MLE's are the best. The second best intervalsare those obtained through pivotal methods or variance stabilization transformation. The third group of intervals is that which depends on the central limit theorem when λ is known. There are two intervals which proved to be unacceptable under any criterion.     

Initial conditions estimation for improving forecast accuracy in exponential smoothing

In this paper we analyze the importance of initial conditions in exponential smoothing models on forecast errors and prediction intervals. We work with certain exponential smoothing models, namely Holt’s additive linear and Gardner’s damped trend. We study some probability properties of those models, showing the influence of the initial conditions on the forecast, which highlights the importance of obtaining accurate estimates of initial conditions. Using the linear heteroscedastic modeling approach, we show how to obtain the joint estimation of initial conditions and smoothing parameters through maximum likelihood via box-constrained nonlinear optimization. Point-wise forecasts of future values and prediction intervals are computed under normality assumptions on the stochastic component. We also propose an alternative formulation of prediction intervals in order to obtain an empirical coverage closer to their nominal values; that formulation adds an additional term to the standard formulas for the estimation of the error variance. We illustrate the proposed approach by using the yearly data time-series from the M3-Competition.     

Admissibility of prediction intervals

Admissibility of prediction intervals is considered in a specified family. It is shown that the best invariant prediction interval is strongly admissible in a location family and in a scale family. Though the similar result has not been obtained for a location and scale family, the best invariant prediction interval for a normal distribution is shown to be weakly admissible.     

Bootstrap prediction intervals in beta regressions

We address the issue of constructing prediction intervals for responses that assume values in the standard unit interval, \((0,1)\) . The response is modeled using the class of beta regression models and we introduce percentile and \(\hbox {BC}_a\) (bias-corrected and accelerated) bootstrap prediction intervals. We present Monte Carlo evidence on the finite sample behavior of such intervals. An empirical application is presented and discussed.     

Estimation of a time series model from unequally spaced data

A process generated by a stochastic differential equation driven by pure noise is sampled at irregular intervals. A model for the sampled sequence is deduced. We describe a maximum likelihood procedure for estimating the parameters and establish the strong consistency and asymptotic normality of the estimates. The use of the model in prediction is considered. Simplifications in the case of periodic sampling are explored.     

Multivariate time series modeling,estimation and prediction of mortalities

We introduce a mixed regression model for mortality data which can be decomposed into a deterministic trend component explained by the covariates age and calendar year, a multivariate Gaussian time series part not explained by the covariates, and binomial risk. Data can be analyzed by means of a simple logistic regression model when the multivariate Gaussian time series component is absent and there is no overdispersion. In this paper we rather allow for overdispersion and the mixed regression model is fitted to mortality data from the United States and Sweden, with the aim to provide prediction and intervals for future mortality and annuity premium, as well as smoothing historical data, using the best linear unbiased predictor. We find that the form of the Gaussian time series has a large impact on the width of the prediction intervals, and it poses some new questions on proper model selection.     

On age-period-cohort parametric mortality rate projections

An enhanced version of the Lee–Carter modelling approach to mortality forecasting, which has been extended to include an age modulated cohort index in addition to the standard age modulated period index, is described and tested for prediction robustness. Life expectancy and annuity value predictions, at pensioner ages and for various periods are compared, both with and without the age modulated cohort index, for the England & Wales male mortality experience. The simulation of prediction intervals for these indices of interest is discussed in detail.     

Two sample Bayesian prediction intervals for order statistics based on the inverse exponential-type distributions using right censored sample

In this paper, two sample Bayesian prediction intervals for order statistics (OS) are obtained. This prediction is based on a certain class of the inverse exponential-type distributions using a right censored sample. A general class of prior density functions is used and the predictive cumulative function is obtained in the two samples case. The class of the inverse exponential-type distributions includes several important distributions such the inverse Weibull distribution, the inverse Burr distribution, the loglogistic distribution, the inverse Pareto distribution and the inverse paralogistic distribution. Special cases of the inverse Weibull model such as the inverse exponential model and the inverse Rayleigh model are considered.     

Learning temporal nodes Bayesian networks

Temporal Nodes Bayesian Networks (TNBNs) are an alternative to Dynamic Bayesian Networks for temporal reasoning with much simpler and efficient models in some domains. TNBNs are composed of temporal nodes, temporal intervals, and probabilistic dependencies. However, methods for learning this type of models from data have not yet been developed. In this paper, we propose a learning algorithm to obtain the structure and temporal intervals for TNBNs from data. The method consists of three phases: (i) obtain an initial approximation of the intervals, (ii) obtain a structure using a standard algorithm and (iii) refine the intervals for each temporal node based on a clustering algorithm. We evaluated the method with synthetic data from three different TNBNs of different sizes. Our method obtains the best score using a combined measure of interval quality and prediction accuracy, and a competitive structural quality with lower running times, compared to other related algorithms. We also present a real world application of the algorithm with data obtained from a combined cycle power plant in order to diagnose temporal faults.