Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Bücher
Zeitschriften
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
ATZ-Fachkongress

Chassis.tech plus 2024


4 Kongresse in einer Veranstaltung

Treffen Sie in München die Fachleute von Chassis, Lenkung, Bremsen sowie Reifen/Räder.
Erweiterte Suche
Anmelden
nach oben

2016 | Buch

Dual Model Misspecification in Generalized Linear Models with Error in Variables Kapitel lesen Erstes Kapitel lesen

New Developments in Statistical Modeling, Inference and Application

Selected Papers from the 2014 ICSA/KISS Joint Applied Statistics Symposium in Portland, OR

herausgegeben von: Zhezhen Jin, Mengling Liu, Xiaolong Luo

Verlag: Springer International Publishing

Buchreihe : ICSA Book Series in Statistics

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

The papers in this volume represent the most timely and advanced contributions to the 2014 Joint Applied Statistics Symposium of the International Chinese Statistical Association (ICSA) and the Korean International Statistical Society (KISS), held in Portland, Oregon. The contributions cover new developments in statistical modeling and clinical research: including model development, model checking, and innovative clinical trial design and analysis. Each paper was peer-reviewed by at least two referees and also by an editor. The conference was attended by over 400 participants from academia, industry, and government agencies around the world, including from North America, Asia, and Europe. It offered 3 keynote speeches, 7 short courses, 76 parallel scientific sessions, student paper sessions, and social events.

Inhaltsverzeichnis

Frontmatter

Theoretical Development in Statistical Modeling

Frontmatter
Dual Model Misspecification in Generalized Linear Models with Error in Variables
Abstract
We study maximum likelihood estimation of regression parameters in generalized linear models for a binary response with error-prone covariates when the distribution of the error-prone covariate or the link function is misspecified. We revisit the remeasurement method proposed by Huang et al. (Biometrika 93:53–64, 2006) for detecting latent-variable model misspecification and examine its operating characteristics in the presence of link misspecification. Furthermore, we propose a new diagnostic method for assessing assumptions on the link function. Combining these two methods yields informative diagnostic procedures that can identify which model assumption is violated and also reveal the direction in which the true latent-variable distribution or the true link function deviates from the assumed one.
Xianzheng Huang
Joint Analysis of Longitudinal Data and Informative Observation Times with Time-Dependent Random Effects
Abstract
Longitudinal data occur in many fields such as the medical follow-up studies that involve repeated measurements. For their analysis, most existing approaches assume that the observation or follow-up times are independent of the response process either completely or given some covariates. In practice, it is apparent that this may not be true. We present a joint analysis approach that allows the possible mutual correlations that can be characterized by time-dependent random effects. Estimating equations are developed for the parameter estimation and the resulting estimators are shown to be consistent and asymptotically normal.
Yang Li, Xin He, Haiying Wang, Jianguo Sun
A Markov Switching Model with Stochastic Regimes with Application to Business Cycle Analysis
Abstract
Since the publication of Hamilton’s seminal work on Markov switching model, a large amount of its applications have been found in economics and finance. As existing Markov switching models describe the regimes or parameter values in a categorical way, it is restrictive in practical analysis. In this paper, we consider a Markov switching model with stochastic regimes, in which the regimes and model parameters are represented both categorically and continuously. Assuming conjugate priors, we develop closed-form recursive Bayes estimates of the regression parameters, an approximation scheme that has much lower computational complexity and yet are comparable to the Bayes estimates in statistical efficiency, and an expectation-maximization procedure to estimate the unknown hyperparameters. We conduct intensive simulation studies to evaluate the performance of our estimators. We also use our model to analyze the series of the U.S. monthly total nonfarm employee.
Haipeng Xing, Ning Sun, Ying Chen
Direction Estimation in a General Regression Model with Discrete Predictors
Abstract
Consider a general regression model, where the response Y depends on discrete predictors X only through the index \( \boldsymbol{\beta }^{T}\mathbf{X} \). It is well-known that the ordinary least squares (OLS) estimator can recover the underlying direction \( \boldsymbol{\beta } \) exactly if the link function between Y and X is linear. Li and Duan (Ann Stat 17:1009–1052, 1989) showed that the OLS estimator can recover \( \boldsymbol{\beta } \) proportionally if the predictors satisfy the linear conditional mean (LCM) condition. For discrete predictors, we demonstrate that the LCM condition generally does not hold. To improve the OLS estimator in the presence of discrete predictors, we model the conditional mean \( \mathrm{E}(\mathbf{X}\mid \boldsymbol{\beta }^{T}\mathbf{X}) \) as a polynomial function of \( \boldsymbol{\beta }^{T}\mathbf{X} \) and use the central solution space (CSS) estimator. The superior performances of the CSS estimators are confirmed through numerical studies.
Yuexiao Dong, Zhou Yu

New Developments in Trial Design

Frontmatter
Futility Boundary Design Based on Probability of Clinical Success Under New Drug Development Paradigm
Abstract
Statistical significance has been the traditional focus of clinical trial design due to the classic requirement for regulatory approval of a new therapy. However, an increasing emphasis is placed on a medical and payer perspective, where the value of a new therapy is generally measured by the magnitude of treatment effect based on point estimates. It is often the case that the magnitude of point estimates to demonstrate sufficient medical value is larger than that to demonstrate statistical significance. Therefore, a new clinical trial design should take into account both statistical significance and the magnitude of point estimates.
In line with the traditional trial design focus being on statistical significance, traditional futility analysis is designed based on power or conditional power to preserve the probability of achieving statistical significance at the end of a trial. With the additional trial objective for a sufficiently large point estimate, we propose an alternative futility analysis design approach where futility boundaries are selected based on the probability of observing a sufficiently large point estimate of treatment effect. We denote such probability as “probability of clinical success”. Additionally, we define “relative retention rate” of this probability, and propose one futility boundary selection criteria to be 90 % relative retention rate. Via an illustrative example, we have extensively evaluated the operational characteristics of this approach including the conditional probability of clinical success based on the interim data and the probability of correct and incorrect stopping, all of which can/should be taken in consideration for futility boundary selection.
Yijie Zhou, Ruji Yao, Bo Yang, Ramachandran Suresh
Bayesian Modeling of Time Response and Dose Response for Predictive Interim Analysis of a Clinical Trial
Abstract
Bayesian approach has been increasingly applied to various aspects of design and analysis of clinical trials. We present one application concerning an interim futility analysis of a trial. Longitudinal data were collected for a range of the studied doses. Bayesian analysis was first conducted to predict observations at the end of treatment for patients not yet followed through treatment, based on all interim observed data. The predicted data in combination with observed data at the end of treatment were then analyzed using a Bayesian normal dynamic linear model for dose response inference. Summary of the Bayesian analysis was used to aid an interim futility decision.
Ming-Dauh Wang, Dominique A. Williams, Elisa V. Gomez, Jyoti N. Rayamajhi
An ROC Approach to Evaluate Interim Go/No-Go Decision-Making Quality with Application to Futility Stopping in the Clinical Trial Designs
Abstract
Interim analyses can be planned to make Go/No-Go decisions in late phase clinical trials and decision quality is an issue of interest because of the timing of interim analysis is often selected based on empirical experience and thresholds for interim Go/No-Go decisions are determined arbitrarily. There is no systematic research to investigate interrelationship among three commonly used statistical methods for interim decision-making, namely conditional power, predictive power, and predicted confidence interval methods. We used a receiver operating characteristic (ROC) approach to evaluate decision-making quality of these methods and they are proved to be equivalent analytically and verified by simulations. To achieve the pre-specified sensitivity and specificity for Go/No-Go decision-making at interim, the required minimum sample size for interim analysis and the threshold for each of three statistical methods can be systematically determined based on the target design parameters of the clinical trials. The application of the obtained results is given for continuous outcome measures.
Deli Wang, Lu Cui, Lanju Zhang, Bo Yang

Novel Applications and Implementation

Frontmatter
Recent Advancements in Geovisualization, with a Case Study on Chinese Religions
Abstract
Producing high-quality, map-based displays for economic, medical, educational, or any other kind of statistical data with geographic covariates has always been challenging. Either it was necessary to have access to high-end software or one had to do a lot of detailed programming. Recently, R software for linked micromap (LM) plots has been enhanced to handle any available shapefiles from Geographic Information Systems (GIS). Also, enhancements have been made that allow for a fast overlay of various statistical graphs on Google maps. In this article, we provide an overview of the necessary steps to produce such graphs in R, starting with GIS-based data and shapefiles and ending with the resulting graphs in R. We will use data from a study on Chinese religions and society (provided by the China Data Center at the University of Michigan) as a case study for these graphical methods.
Jürgen Symanzik, Shuming Bao, XiaoTian Dai, Miao Shui, Bing She
The Efficiency of Next-Generation Gibbs-Type Samplers: An Illustration Using a Hierarchical Model in Cosmology
Abstract
Supernovae occur when a star’s life ends in a violent thermonuclear explosion, briefly outshining an entire galaxy before fading from view over a period of weeks or months. Because so-called Type Ia supernovae occur only in a particular physical scenario, their explosions have similar intrinsic brightnesses which allows us to accurately estimate their distances. This in turn allows us to constrain the parameters of cosmological models that characterize the expansion history of the universe. In this paper, we show how a cosmological model can be embedded into a Gaussian hierarchical model and fit using observations of Type Ia supernovae. The overall model is an ideal testing ground of new computational methods. Ancillarity-Sufficiency Interweaving Strategy (ASIS) and Partially Collapsed Gibbs (PCG) are effective tools to improve the convergence of Gibbs samplers. Besides using either of them alone, we can combine PCG and/or ASIS along with Metropolis-Hastings algorithm to simplify implementation and further improve convergence. We use four samplers to draw from the posterior distribution of the cosmological hierarchical model, and confirm the efficiency of both PCG and ASIS. Furthermore, we find that we can gain more efficiency by combining two or more strategies into one sampler.
Xiyun Jiao, David A. van Dyk, Roberto Trotta, Hikmatali Shariff
Dynamic Spatial Pattern Recognition in Count Data
Abstract
This study explores a Bayesian regression analysis for count data in the presence of spatial and temporal correlations. The contribution is to develop a regression model for count data that provides flexibility in modeling the complexity of zero-inflation, overdispersion, as well as spatial patterns in a dynamic manner. More importantly, it improves the computational efficiency via dimension reduction while handling the high-dimensional correlation structure in the data. The proposed model is applied to the survey data by the Northeast Fisheries Sciences Center (NEFSC) for estimation and prediction of the Atlantic cod in the Gulf of Maine—Georges Bank region. Both zero-inflated Poisson and negative binomial models are fitted. Model comparison shows the improvement in model fitting with consideration in the spatial-temporal correlation as well as the overdispersion in the count data.
Xia Wang, Ming-Hui Chen, Rita C. Kuo, Dipak K. Dey
Bias-corrected Estimators of Scalar Skew Normal
Abstract
One problem of a skew normal model is the difficulty in estimating the shape parameter, for which the maximum likelihood estimate may be infinite when sample size is moderate. The existing estimators suffer from large bias even for moderate size samples. In this paper, we proposed five estimators of the shape parameter for a scalar skew normal model, either by bias correction method or by solving a modified score equation. Simulation studies show that except bootstrap estimator, the proposed estimators have smaller bias compared to those estimators in literature for small and moderate samples.
Guoyi Zhang, Rong Liu
Metadaten
Titel
New Developments in Statistical Modeling, Inference and Application
herausgegeben von
Zhezhen Jin
Mengling Liu
Xiaolong Luo
Copyright-Jahr
2016
Electronic ISBN
978-3-319-42571-9
Print ISBN
978-3-319-42570-2
DOI
https://doi.org/10.1007/978-3-319-42571-9

Premium Partner

    Bildnachweise