Search Results

Concluding Remarks

Guo Shenyang

in Survival Analysis

This chapter presents some concluding thoughts from the author. It comments on criticisms of studies using survival analysis, and on directions for future development. It calls for the continuous ... More

This chapter presents some concluding thoughts from the author. It comments on criticisms of studies using survival analysis, and on directions for future development. It calls for the continuous development and improvement of approaches to handling clustered event times, the need to solidify the statistical theories of survival analysis by connecting the existing models (e.g., the Kaplan–Meier estimator and the Cox proportional hazards model) to the study of counting process and martingale theory, and a more innovative and wider application of the advanced survival models to solving research problems outside the field of biomedics.Less

Computing Software Packages for Survival Analysis

Guo Shenyang

in Survival Analysis

All commercial software packages offer procedures for survival analysis. This chapter provides an overview to highlight key issues in programming with SAS, SPSS, and Stata.

Introduction

Guo Shenyang

in Survival Analysis

This introductory chapter begins with a description of survival analysis. It then discusses why and when survival analysis is needed, and the significance of conducting survival analysis. An overview ... More

This introductory chapter begins with a description of survival analysis. It then discusses why and when survival analysis is needed, and the significance of conducting survival analysis. An overview of the subsequent chapters is presented.Less

Profit scoring and dynamic models

Lyn C. Thomas

in Consumer Credit Models: Pricing, Profit and Portfolios

This chapter begins by reviewing the role of behavioural scoring and risk/reward matrices in the way a lender manages borrowers. It points out that current methods do not allow for the future changes ... More

This chapter begins by reviewing the role of behavioural scoring and risk/reward matrices in the way a lender manages borrowers. It points out that current methods do not allow for the future changes in customer behaviour nor do they optimize the expected profit. It then shows how two approaches — Markov chain models and survival analysis ideas — can lead to dynamic profitability based models. In particular, the extension of Markov chains to Markov decision processes allows one to optimize decisions such as how to adjust the credit limit. In survival analysis, the proportional hazard models lead to hazard scores which play the same role as credit scores but work on all future performance periods. Moreover, the ideas of competing risk in survival analysis allows one to build profitability models that allow for default, prepayment, and attrition.Less

Key Concepts and Descriptive Approaches

Shenyang Guo

in Survival Analysis

This chapter reviews key concepts of survival analysis, two descriptive methods (i.e., the life-table approach and the Kaplan–Meier estimate of survivor function), and graphic approaches. These ... More

This chapter reviews key concepts of survival analysis, two descriptive methods (i.e., the life-table approach and the Kaplan–Meier estimate of survivor function), and graphic approaches. These methods are typically employed at the beginning stage of a quantitative inquiry of time-to-event data, although almost all concepts described in the chapter are important in understanding the entire survival analysis method.Less

Event Time Analysis

Ezra Susser, Sharon Schwartz, Alfredo Morabia, and Evelyn J. Bromet

in Psychiatric Epidemiology: Searching for the Causes of Mental Disorders

This chapter describes statistical methods for taking account of unequal attrition, that is, different follow-up times across exposed and unexposed groups in a cohort study. It considers methods for ... More

This chapter describes statistical methods for taking account of unequal attrition, that is, different follow-up times across exposed and unexposed groups in a cohort study. It considers methods for analyzing time to an event as an outcome in its own right. The chapter defines the outcome as the time until disease onset. From this perspective, a nondiseased subject in the study is not a control as in a case-control study, but someone who has not yet developed the disease. To analyze disease incidence data from this perspective, methods that come under the interchangeable headings of survival analysis, time-to-event analysis, or failure-time analysis are used.Less

Analysis of DNA methylation in cancer

Fabian Model, Jörn Lewin, Catherine Lofton-Day, and Gunter Weiss

in Statistics and Informatics in Molecular Cancer Research

This chapter gives an overview of algorithms and statistical methods used for measuring and analyzing DNA methylation in cancer research. It starts with a short introduction into the biology of DNA ... More

This chapter gives an overview of algorithms and statistical methods used for measuring and analyzing DNA methylation in cancer research. It starts with a short introduction into the biology of DNA methylation and its role in cancer. This is followed by an overview of measurement technologies and a detailed description of data pre-processing algorithms for normalization and calibration of direct bisulphite DNA sequencing and DNA methylation microarray measurements. The second part of the chapter explores some typical examples of DNA methylation data analysis in cancer research and diagnostics: the classification of tumour tissue samples, the detection of cancer in plasma samples, and the tissue based prediction of tumour recurrence.Less

Event History Data

Ludwig Fahrmeir and Thomas Kneib

in Bayesian Smoothing and Regression for Longitudinal, Spatial and Event History Data

This chapter extends Bayesian approaches for smoothing and regression developed in previous chapters to regression models for survival and event history data with structured additive predictors. This ... More

This chapter extends Bayesian approaches for smoothing and regression developed in previous chapters to regression models for survival and event history data with structured additive predictors. This allows for the inclusion of nonlinear time-varying effects and flexible covariate effects, spatial effects, and random effects in addition to common linear predictors and to estimate them simultaneously based on full or empirical Bayes inference. Alternative approaches and other model types are outlined in Section 6.6.Less

Survival Data: Estimation Of Risk

Steve Selvin

in Statistical Analysis of Epidemiologic Data

Therapeutic trials usually involve individuals observed over time where the outcome might be death or the recurrence of a specific disease. A basic characteristic that separates survival data from ... More

Therapeutic trials usually involve individuals observed over time where the outcome might be death or the recurrence of a specific disease. A basic characteristic that separates survival data from other kinds of data is that the outcome is frequently not observed for all study subjects. Special statistical methods exist to compensate for the incomplete nature of this kind of data. The theory and application of these methods is called survival analysis. This chapter discusses two general approaches to survival analysis — one parametric and one nonparametric.Less

Evaluation of Risks, Decision Making, and Outcomes

Leslie R. Martin, Kelly B. Haskard-Zolnierek, and M. Robin DiMatteo

in Health Behavior Change and Treatment Adherence: Evidence-based Guidelines for Improving Healthcare

Few people conduct a truly thorough and thoughtful evaluation of the evidence before they make a health-related decision. This chapter describes and evaluates strategies for making decisions based on ... More

Few people conduct a truly thorough and thoughtful evaluation of the evidence before they make a health-related decision. This chapter describes and evaluates strategies for making decisions based on empirical evidence. It then overviews the elements that are needed in order to understand health and medical risk information (e.g., Bayesian methods, odds ratios, risk ratios, survival analyses, and hazard ratios) as that information is typically presented in medical journals, scientific articles, news reports, and advertisements. The relative power of aggregated data (through meta-analysis) is also discussed. Evidence supporting the crucial role of the patient in decision making, and specific tools that can be used in decision making are presented (e.g., decision trees, PREPARED™), is reviewed.Less

Survival Data: Proportional Hazards Model

Steve Selvin

in Statistical Analysis of Epidemiologic Data

The success of a model-based approach depends on choosing a model that accurately reflects the relationships within the data. This choice requires knowledge of the statistical properties of the model ... More

The success of a model-based approach depends on choosing a model that accurately reflects the relationships within the data. This choice requires knowledge of the statistical properties of the model and a clear understanding of the phenomenon being investigated. One of the many useful models applied to survival data is the proportional hazards model. This chapter describes this model in simple terms, illustrating its properties and providing insight into the process of analyzing survival experience data using statistical modeling techniques.Less

The Parametric Models

Guo Shenyang

in Survival Analysis

This chapter presents a review of parametric models (exponential model, Weibull model, piecewise exponential model). Parametric models are the method from which the contemporary survival analysis ... More

This chapter presents a review of parametric models (exponential model, Weibull model, piecewise exponential model). Parametric models are the method from which the contemporary survival analysis originates; particularly, the model is an extension of the traditional ordinary least squares (OLS) regression; understanding the main features of parametric models helps users understand the fundamental concepts of survival analysis as well as important statistical concepts in general. The parametric models can also be used to solve unique problems, such as handling left-hand and interval censorings; one of interesting parametric models is the piecewise exponential model, which is widely applied by social, health, and behavioral scientists.Less

Survival analysis

Bendix Carstensen

in Epidemiology with R

This chapter describes survival analysis. Survival analysis concerns data where the outcome is a length of time, namely the time from inclusion in the study (such as diagnosis of some disease) till ... More

This chapter describes survival analysis. Survival analysis concerns data where the outcome is a length of time, namely the time from inclusion in the study (such as diagnosis of some disease) till death or some other event — hence the term 'time to event analysis', which is also used. There are two primary targets normally addressed in survival analysis: survival probabilities and event rates. The chapter then looks at the life table estimator of survival function and the Kaplan–Meier estimator of survival. It also considers the Cox model and its relationship with Poisson models, as well as the Fine–Gray approach to competing risks.Less

Cities, Redistribution, and Regime Survival

Jeremy L. Wallace

in Cities and Stability: Urbanization, Redistribution, and Regime Survival in China

The third chapter examines the survival patterns of authoritarian regimes after World War II, demonstrating the dangers cities pose to such nondemocratic regimes. Simple attempts to buy off urbanites ... More

The third chapter examines the survival patterns of authoritarian regimes after World War II, demonstrating the dangers cities pose to such nondemocratic regimes. Simple attempts to buy off urbanites temporarily sustain but ultimately undermine regimes. Where largest cities are more dominant, collective action is more common. Data from 435 nondemocratic regimes in over 100 countries confirm three principle hypotheses. First is the danger of concentration hypothesis: larger cities and higher levels of urban concentration negatively affect regime survival. Second is the induced concentration hypothesis: urban bias should induce additional migration to favored large cities. Third is the Faustian Bargain hypothesis: dominant cities can be stabilized by urban bias today but can grow to be overwhelming and undermine regime survival if not held in check. These findings are robust across numerous specifications, the inclusion of control variables, and for subsets of the data.Less

Foreign Aid and Democratic Development in Africa

Simone Dietrich and Joseph Wright

in Democratic Trajectories in Africa: Unravelling the Impact of Foreign Aid

Over the past two decades, donors increasingly linked foreign aid to democracy objectives in sub-Saharan Africa. Yet systematic research on this topic typically focuses on how aid influences ... More

Over the past two decades, donors increasingly linked foreign aid to democracy objectives in sub-Saharan Africa. Yet systematic research on this topic typically focuses on how aid influences democratic transitions. This study investigates whether and how foreign aid affects the process of democratic consolidation in sub-Saharan Africa by examining two potential mechanisms: (1) the use of aid as leverage to buy political reform, and (2) investment in the opposition. We test these mechanisms using five dependent variables that capture different aspects of democratic consolidation. Using survival analysis for the period from 1991 to 2008, we find that democracy and governance aid has a consistently positive effect on democratic consolidation. Economic aid, on the other hand, has no effect on democratic consolidationLess

Factors of Post-Conflict Peace Stabilization

Richard Caplan

in Measuring Peace: Principles, Practices, and Politics

This chapter (co-written with Anke Hoeffler) seeks to identify factors that contribute to post-conflict peace stabilization based on a quantitative analysis using duration (survival) analysis and a ... More

This chapter (co-written with Anke Hoeffler) seeks to identify factors that contribute to post-conflict peace stabilization based on a quantitative analysis using duration (survival) analysis and a qualitative analysis examining the peace consolidation process in six conflict-affected countries. Duration analysis, a statistical method, allows us to analyse the duration of peace. The hazard rate—the rate at which peace ends—can be modelled as a function of various co-variates, such as economic growth, aid, elections, military personnel and expenditure, regional autonomy, etc. The country case studies provide more detailed information on how some countries achieved lasting peace while others failed. The country cases that are included in this analysis are: Burundi, El Salvador, Liberia, Nepal, Sierra Leone, and Timor-Leste (East Timor).Less

Patterns of New Party Persistence and Sustainability in Seventeen Democracies

Nicole Bolleyer

in New Parties in Old Party Systems: Persistence and Decline in Seventeen Democracies

This chapter examines which factors shape new parties’ organizational persistence looking at new parties’ overall life cycles and of their electoral sustainability, i.e. whether they manage to ... More

This chapter examines which factors shape new parties’ organizational persistence looking at new parties’ overall life cycles and of their electoral sustainability, i.e. whether they manage to re-enter parliament after their national breakthrough. More particularly, it assesses the importance of party origin. As theoretically expected, the statistical analyses show that rooted formations are not only more likely to live longer (persistence) but also more likely to maintain a presence on the national level after its national breakthrough by assuring repeated parliamentary entry (sustainability). Its concluding section explains how the following in-depth case studies will complement these findings by bringing party agency into the picture and examine claims linked to the leadership-structure dilemma as specified in Chapter 3. It further identifies systematic linkages between parties’ origins and programmatic profiles. The resulting ‘clusters’ or ‘new party families’ provide the basic logic along which the following qualitative case studies will be organized.Less

Mixture models for overdispersed data

Jonathan R. Rhodes

in Ecological Statistics: Contemporary theory and application

Ecological data often do not conform to the assumptions of standard probability distributions and this has important implications for the validity of statistical inference. A common reason for this ... More

Ecological data often do not conform to the assumptions of standard probability distributions and this has important implications for the validity of statistical inference. A common reason for this is that the variability of ecological data is often much higher than can be accounted for by the standard probability distributions that underpin most statistical inference in ecology. This leads to an underestimation of variances and bias in statistical tests unless the overdispersion is accounted for. Consequently, having methods for dealing with overdispersion is an essential component of the ecologist’s statistical toolbox. This chapter introduces statistical methods known as mixture models that can deal with overdispersion. Mixture models are powerful because not only can they account for overdispersion, but they can also help to identify the actual ecological or observation processes that drive overdispersion. The chapter begins by discussing the causes and consequences of overdispersion in ecological data and how overdispersion can be identified. Mixture models are then described and illustrated using two different case studies from survival analysis and the analysis of population abundance. The chapter ends with a discussion of some of the limitations of mixture models and pitfalls to look out for.Less

Introduction

Agar Brugiavini and Guglielmo Weber

in Longer-term Consequences of the Great Recession on the Lives of Europeans

This chapter introduces the reader to the topics covered in the book. It explains in what sense and to what extent the analysis of individual-level survey data can be informative on the way the ... More

This chapter introduces the reader to the topics covered in the book. It explains in what sense and to what extent the analysis of individual-level survey data can be informative on the way the current recession will affect the lives of Europeans in the near and distant future, and discusses the methodology used in the three main chapters that present novel results from European survey data.Less

Anatomy of Financial Hardship

Agar Brugiavini and Guglielmo Weber

in Longer-term Consequences of the Great Recession on the Lives of Europeans

This chapter uses life-history data covering a number of European countries (SHARELIFE) to analyse what makes people likely to fall into financial hardship. The analysis highlights the all-important ... More

This chapter uses life-history data covering a number of European countries (SHARELIFE) to analyse what makes people likely to fall into financial hardship. The analysis highlights the all-important role of such shocks as divorce, birth of children and their nest-leaving, or the inception of poor health and bereavement, but also suggests the importance of loss of job and retirement (particularly if brought about by poor health). Inflation and recession episodes also play a direct role, but this is much smaller for those who are not otherwise affected. Also, for those who are in financial hardship, we investigate the factors that make people more likely to recover earlier, and find important differences across countries characterized by different welfare systems.Less