Approved for public release; distribution is unlimited / The determination of a system's life distribution usually requires the synthesis of a mixture of system survival modes. In order to alleviate the normal non-trivial calculations, this paper presents the concept of a reliability shorthand. After describing the possible ways a system can survive a mission, the practitioner of this shorthand can use stock formulas to obtain a system's survival function. Then simple insertion of the failure rates of the system's components into the known equations results in the system's reliability. Simple examples show the convenience of this shorthand. The TI-59 is demonstrated to be a useful tool; adequate to implement the methodology. / http://archive.org/details/introductiontore00repi / Lieutenant Commander, United States Navy

Kevlar materials offer excellent tensile and thermal properties but can rapidly degrade under exposure to hot and humid environmental conditions. Currently Kevlar fiber's survival probability comes from a single filament test. Unfortunately, the single filament test is a tedious process and prone to operator bias, leading to inaccurate survival function that does not represent the actual survival function. This research aims to validate the fiber bundle test to replace the single filament test in extracting Kevlar’s survival function. Another important aspect is determining the factors that cause the fiber to lose its properties. This research also aims to determine the factors that degrade Kevlar fibers and those factors’ combined effects on degrading the KM2 fiber. This information is essential for safety factor design when exposure to these environmental factors would cause the Kevlar KM2 to fail prematurely. Results from experimental data and analysis indicate that the fiber bundle test is a good replacement for single filament tests and estimation techniques can determine the bundle Weibull parameters. Furthermore, the survival function for treated fibers is better if the bundle is lubricated. The accelerated aging experiments show that accelerated aging is possible with combined temperature and moisture. Kevlar KM2 bundle conditioned at 270 °C and 150 g water for 3 hours lost over 95% of its breaking strength. This is comparable to Kevlar bundles treated for over 500 hours in 250 °C or treated for over 100 days in 100% relative humidity environment at 80 °C found in literature.

When there are a large number of predictors and few observations, building a regression model to explain the behavior of a response variable such as a patient's medical condition is very challenging. This is a "p ≫n " variable selection problem encountered often in modern applied statistics and data mining. Chapter one of this thesis proposes a rigorous procedure which groups predictors into clusters of "highly-correlated" variables, selects a representative from each cluster, and uses a subset of the representatives for regression modeling. The proposed Penalized method based on Representatives (PR) extends the Lasso for the p ≫ n data and highly correlated variables, to build a sparse model practically interpretable and maintain prediction quality. Moreover, we provide the PR-Sequential Grouped Regression (PR-SGR) to make computation of the PR procedure efficient. Simulation studies show the proposed method outperforms existing methods such as the Lasso/Lars. A real-life example from a mental health diagnosis illustrates the applicability of the PR-SGR. In the second part of the thesis, we study the analysis of time-to-event data called a gap data when missing time intervals (gaps) possibly happen prior to the first observed event time. If a gap occurs prior to the first observed event, then the first observed event may or may not be the first true event. This incomplete knowledge makes the gap data different from the well-studied regular interval censored data. We propose a Non-Parametric Estimate for the Gap data (NPEG) to estimate the survival function for the first true event time, derive its analytic properties and demonstrate its performance in simulations. We also extend the Imputed Empirical Estimating method (IEE), which is an existing nonparametric method for the gap data up to one gap, to handle the gap data with multiple gaps.

A life distribution and its survival function F = 1 - F with finitemean y = /q F(x)dx are said to be HNBUE (HNWUE) if F(x)dx &lt; (&gt;)U exp(-t/y) for t &gt; 0. The major part of this thesis deals with the classof HNBUE (HNWUE) life distributions. We give different characterizationsof the HNBUE (HNWUE) property and present bounds on the moments and on thesurvival function F when this is HNBUE (HNWUE). We examine whether theHNBUE (HNWUE) property is preserved under some reliability operations andstudy some test statistics for testing exponentiality against the HNBUE(HNWUE) property.The HNBUE (HNWUE) property is studied in connection with shock models.Suppose that a device is subjected to shocks governed by a counting processN = {N(t): t &gt; 0}. The probability that the device survives beyond t isthen00H(t) = S P(N(t)=k)P, ,k=0where P^ is the probability of surviving k shocks. We prove that His HNBUE (HNWUE) under different conditions on N and * ^orinstance we study the situation when the interarrivai times between shocksare independent and HNBUE (HNWUE).We also study the Pure Birth Shock Model, introduced by A-Hameed andProschan (1975), and prove that H is IFRA and DMRL under conditions whichdiffer from those used by A-Hameed and Proschan.Further we discuss relationships between the total time on test transformHp^(t) = /q ^F(s)ds , where F \t) = inf { x: F(x) &gt; t}, and differentclasses of life distributions based on notions of aging. Guided by propertiesof we suggest test statistics for testing exponentiality agains t IFR,IFRA, NBUE, DMRL and heavy-tailedness. Different properties of these statisticsare studied.Finally, we discuss some bivariate extensions of the univariate properties NBU, NBUE, DMRL and HNBUE and study some of these in connection with bivariate shock models. / <p>There are some occurring misspellings in the formulas in the abstract on this webpage. Read the abstract in the full-text document for correct spelling in formulas, see the downloadable file.</p> / digitalisering@umu

Aalen’s additive hazards model has gained increasing attention in recently years because it model all covariate effects as time-varying. In this thesis, our goal is to explore the application of Aalen’s model in assessing treatment effect at a given time point with varying covariate effects. First, based on Aalen’s model, we utilize the direct adjustment method to obtain the adjusted survival of a treatment and comparing two direct adjusted survivals, with univariate survival data. Second, we focus on application of Aalen’s model in the setting of competing risks data, to assess treatment effect on a particular type of failure. The direct adjusted cumulative incidence curve is introduced. We further construct the confidence interval of the difference between two direct adjusted cumulative incidences, to compare two treatments on one risk.

This work introduces nonparametric models which are used in time to event data analysis. It is focused on applying these methods in medicine where it is called survival analysis. The basic techniques and problems, which can appear in survival analysis, are presented and explained here. The Kaplan -- Meier estimator of survival function is discussed in the main part. This is the most frequented method used for estimating the survival function in patients who have undergone a specific treatment. The Kaplan -- Meier estimator is also a common device in the statistical packets. In addition to estimation of survival function, the estimation of hazard function and cumulative hazard function is presented. The hazard function shows the intensity of an individual experiencing the particular event in a short time period. Special problems occur when analyzing time to event data. A distinctive feature, often present in such data, is known as censoring. That is the situation when the individual does not experience the event of interest at the time of study. The thesis covers also an empiric part, where the results of an analysis of patients with the larynx carcinoma diagnosis are shown. These patients were treated in a hospital located in České Budějovice. This analysis is based on a theory presented in the previous chapters.

No presente trabalho são propostos dois modelos para dados censurados baseados na mistura da distribuição geométrica e na distribuição Burr XII considerando duas ativações latentes, máximo e mínimo. A distribuição Burr XII tem três parâmetros e é uma generalização da distribuição log-logística. Por sua vez a distribuição Burr XII Geométrica tipo I e tipo II tem quatro parâmetros e são generalizações da distribuição Burr XII relacionados as ativações latentes do mínimo e máximo respectivamente. Foram apresentadas algumas propriedades das duas novas distribuições tais como momentos, assimetria, curtose, função geradora de momentos e desvio médio. Além disso, foi intriduzido os modelos de regressão correspondentes, log Burr XII Geométrica tipo I e log Burr XII Geométrica tipo II. Adicionalmente foi desenvolvido um modelo de sobrevivência com fração de cura assumindo que o número de causas competitivas do evento de interesse segue a distribuição geométrica e o tempo do evento segue a distribuição Burr XII. Para todos os modelos desenvolvidos foi utilizado o método da máxima verossimilhança para estimar os parâmetros, que possibilita a construção de intervalos de confiança e testes de hipóteses. Por fim, são apresentadas três aplicações para ilustrar os modelos propostos. / In this paper are proposed two models for censored data based on the mixture of geometric distribution and Burr XII distribution considering two latent activations, maximum and minimum. The Burr XII distribution has three parameters and is a generalization of the log-logistic distribution. On the other hand Burr XII Geometric type I distribution and type II has four parameters and are a generalization of the Burr XII distribution related to minimum and maximum activations respectively. It were presented some properties of the news distributions such as moments, skewness, kurtosis, moment generating function and mean deviation. Furthermore, it was introduced two regression models, the log Burr XII Geometric type I and the log Burr XII Geometric type II. Additionally a new cure rate survival was formulated by assuming that the number of competing causes of the event of interest has the geometric distribution and the time to this event follows Burr XII distribution. For all models was developed the maximum likelihood method to estimate the parameters, which allows the construction of confidence intervals and hypothesis testing. Finally, three applications are presented to illustrate the proposed models.

In the dissertation, we study the statistical evaluation of treatment comparisons by evaluating the relative comparison of survival experiences between two treatment groups. We construct confidence interval and simultaneous confidence bands for the ratio and odds ratio of two survival functions through both parametric and nonparametric approaches.We first construct empirical likelihood confidence interval and simultaneous confidence bands for the odds ratio of two survival functions to address small sample efficacy and sufficiency. The empirical log-likelihood ratio is developed, and the corresponding asymptotic distribution is derived. Simulation studies show that the proposed empirical likelihood band has outperformed the normal approximation band in small sample size cases in the sense that it yields closer coverage probabilities to chosen nominal levels.Furthermore, in order to incorporate prognostic factors for the adjustment of survival functions in the comparison, we construct simultaneous confidence bands for the ratio and odds ratio of survival functions based on both the Cox model and the additive risk model. We develop simultaneous confidence bands by approximating the limiting distribution of cumulative hazard functions by zero-mean Gaussian processes whose distributions can be generated through Monte Carlo simulations. Simulation studies are conducted to evaluate the performance for proposed models. Real applications on published clinical trial data sets are also studied for further illustration purposes.In the end, the population attributable fraction function is studied to measure the impact of risk factors on disease incidence in the population. We develop semiparametric estimation of attributable fraction functions for cohort studies with potentially censored event time under the additive risk model.

In the dissertation, we study the statistical evaluation of treatment comparisons by evaluating the relative comparison of survival experiences between two treatment groups. We construct confidence interval and simultaneous confidence bands for the ratio and odds ratio of two survival functions through both parametric and nonparametric approaches.We first construct empirical likelihood confidence interval and simultaneous confidence bands for the odds ratio of two survival functions to address small sample efficacy and sufficiency. The empirical log-likelihood ratio is developed, and the corresponding asymptotic distribution is derived. Simulation studies show that the proposed empirical likelihood band has outperformed the normal approximation band in small sample size cases in the sense that it yields closer coverage probabilities to chosen nominal levels.Furthermore, in order to incorporate prognostic factors for the adjustment of survival functions in the comparison, we construct simultaneous confidence bands for the ratio and odds ratio of survival functions based on both the Cox model and the additive risk model. We develop simultaneous confidence bands by approximating the limiting distribution of cumulative hazard functions by zero-mean Gaussian processes whose distributions can be generated through Monte Carlo simulations. Simulation studies are conducted to evaluate the performance for proposed models. Real applications on published clinical trial data sets are also studied for further illustration purposes.In the end, the population attributable fraction function is studied to measure the impact of risk factors on disease incidence in the population. We develop semiparametric estimation of attributable fraction functions for cohort studies with potentially censored event time under the additive risk model.

No presente trabalho são propostos dois modelos para dados censurados baseados na mistura da distribuição geométrica e na distribuição Burr XII considerando duas ativações latentes, máximo e mínimo. A distribuição Burr XII tem três parâmetros e é uma generalização da distribuição log-logística. Por sua vez a distribuição Burr XII Geométrica tipo I e tipo II tem quatro parâmetros e são generalizações da distribuição Burr XII relacionados as ativações latentes do mínimo e máximo respectivamente. Foram apresentadas algumas propriedades das duas novas distribuições tais como momentos, assimetria, curtose, função geradora de momentos e desvio médio. Além disso, foi intriduzido os modelos de regressão correspondentes, log Burr XII Geométrica tipo I e log Burr XII Geométrica tipo II. Adicionalmente foi desenvolvido um modelo de sobrevivência com fração de cura assumindo que o número de causas competitivas do evento de interesse segue a distribuição geométrica e o tempo do evento segue a distribuição Burr XII. Para todos os modelos desenvolvidos foi utilizado o método da máxima verossimilhança para estimar os parâmetros, que possibilita a construção de intervalos de confiança e testes de hipóteses. Por fim, são apresentadas três aplicações para ilustrar os modelos propostos. / In this paper are proposed two models for censored data based on the mixture of geometric distribution and Burr XII distribution considering two latent activations, maximum and minimum. The Burr XII distribution has three parameters and is a generalization of the log-logistic distribution. On the other hand Burr XII Geometric type I distribution and type II has four parameters and are a generalization of the Burr XII distribution related to minimum and maximum activations respectively. It were presented some properties of the news distributions such as moments, skewness, kurtosis, moment generating function and mean deviation. Furthermore, it was introduced two regression models, the log Burr XII Geometric type I and the log Burr XII Geometric type II. Additionally a new cure rate survival was formulated by assuming that the number of competing causes of the event of interest has the geometric distribution and the time to this event follows Burr XII distribution. For all models was developed the maximum likelihood method to estimate the parameters, which allows the construction of confidence intervals and hypothesis testing. Finally, three applications are presented to illustrate the proposed models.