Global ETD Search

81	Statistical analysis of clinical trial data using Monte Carlo methods Han, Baoguang 11 July 2014 (has links) Indiana University-Purdue University Indianapolis (IUPUI) / In medical research, data analysis often requires complex statistical methods where no closed-form solutions are available. Under such circumstances, Monte Carlo (MC) methods have found many applications. In this dissertation, we proposed several novel statistical models where MC methods are utilized. For the first part, we focused on semicompeting risks data in which a non-terminal event was subject to dependent censoring by a terminal event. Based on an illness-death multistate survival model, we proposed flexible random effects models. Further, we extended our model to the setting of joint modeling where both semicompeting risks data and repeated marker data are simultaneously analyzed. Since the proposed methods involve high-dimensional integrations, Bayesian Monte Carlo Markov Chain (MCMC) methods were utilized for estimation. The use of Bayesian methods also facilitates the prediction of individual patient outcomes. The proposed methods were demonstrated in both simulation and case studies. For the second part, we focused on re-randomization test, which is a nonparametric method that makes inferences solely based on the randomization procedure used in clinical trials. With this type of inference, Monte Carlo method is often used for generating null distributions on the treatment difference. However, an issue was recently discovered when subjects in a clinical trial were randomized with unbalanced treatment allocation to two treatments according to the minimization algorithm, a randomization procedure frequently used in practice. The null distribution of the re-randomization test statistics was found not to be centered at zero, which comprised power of the test. In this dissertation, we investigated the property of the re-randomization test and proposed a weighted re-randomization method to overcome this issue. The proposed method was demonstrated through extensive simulation studies. Numerical analysis -- Data processing Random variables -- Research Bayesian statistical decision theory Mortality -- Mathematical models Clinical trials -- Statistical methods Medical statistics Biology -- Data processing Probabilities -- Data processing
82	Joint models for longitudinal and survival data Yang, Lili 11 July 2014 (has links) Indiana University-Purdue University Indianapolis (IUPUI) / Epidemiologic and clinical studies routinely collect longitudinal measures of multiple outcomes. These longitudinal outcomes can be used to establish the temporal order of relevant biological processes and their association with the onset of clinical symptoms. In the first part of this thesis, we proposed to use bivariate change point models for two longitudinal outcomes with a focus on estimating the correlation between the two change points. We adopted a Bayesian approach for parameter estimation and inference. In the second part, we considered the situation when time-to-event outcome is also collected along with multiple longitudinal biomarkers measured until the occurrence of the event or censoring. Joint models for longitudinal and time-to-event data can be used to estimate the association between the characteristics of the longitudinal measures over time and survival time. We developed a maximum-likelihood method to joint model multiple longitudinal biomarkers and a time-to-event outcome. In addition, we focused on predicting conditional survival probabilities and evaluating the predictive accuracy of multiple longitudinal biomarkers in the joint modeling framework. We assessed the performance of the proposed methods in simulation studies and applied the new methods to data sets from two cohort studies. / National Institutes of Health (NIH) Grants R01 AG019181, R24 MH080827, P30 AG10133, R01 AG09956. joint models longitudinal data survival data bivariate change point models prediction Bayesian method EM algorithm Biologically-inspired computing Probability measures Expectation-maximization algorithms Failure time data analysis Numerical analysis -- Data processing Clinical trials -- Statistical methods
83	Sur les familles des lois de fonction de hasard unimodale : applications en fiabilité et analyse de survie Saaidia, Noureddine 24 June 2013 (has links) En fiabilité et en analyse de survie, les distributions qui ont une fonction de hasard unimodale ne sont pas nombreuses, qu'on peut citer: Gaussienne inverse ,log-normale, log-logistique, de Birnbaum-Saunders, de Weibull exponentielle et de Weibullgénéralisée. Dans cette thèse, nous développons les tests modifiés du Chi-deux pour ces distributions tout en comparant la distribution Gaussienne inverse avec les autres. Ensuite nousconstruisons le modèle AFT basé sur la distribution Gaussienne inverse et les systèmes redondants basés sur les distributions de fonction de hasard unimodale. / In reliability and survival analysis, distributions that have a unimodalor $\cap-$shape hazard rate function are not too many, they include: the inverse Gaussian,log-normal, log-logistic, Birnbaum-Saunders, exponential Weibull and power generalized Weibulldistributions. In this thesis, we develop the modified Chi-squared tests for these distributions,and we give a comparative study between the inverse Gaussian distribution and the otherdistributions, then we realize simulations. We also construct the AFT model based on the inverseGaussian distribution and redundant systems based on distributions having a unimodal hazard ratefunction. Distribution Gaussienne inverse, Distribution log-normale, Distribution log-logistique, Distribution de Birnbaum-Saunders, Distribution de Weibull exponentielle, Distribution de Weibull généralisée, Test du Chi-deux, Test modifié du Chi-deux, Fiabilité, Analyse de survie, Modèle de Sedyakin, Modèle AFT, Systèmes redondants, Intervalle de confiance, Temps de panne, Maximum de vraisemblance, Estimation, Classes de Neyman-Pearson. Inverse Gaussian distribution, Log-normal distribution, Log-logistic distribution, Birnbaum-Saunders Distribution , Exponential Weibull distribution, Power generalized Weibull distribution, Chi-square test, Modified Chi-squared, Reliability, Survival analysis, Sedyakin's model, AFT model, Redundant system, Confidence interval, Failure time, Maximum likelihood , Estimation, Neyman-Pearson classes Maximum de vraisemblance,

Page generated in 0.0678 seconds