Muutke küpsiste eelistusi

Introducing Survival and Event History Analysis [Pehme köide]

3.83/5 (8 hinnangut Goodreads-ist)
  • Formaat: Paperback / softback, 300 pages, kõrgus x laius: 242x170 mm, kaal: 510 g
  • Ilmumisaeg: 21-Dec-2010
  • Kirjastus: Sage Publications Ltd
  • ISBN-10: 1848601026
  • ISBN-13: 9781848601024
Teised raamatud teemal:
Introducing Survival and Event History AnalysisSuurem pilt
  • Formaat: Paperback / softback, 300 pages, kõrgus x laius: 242x170 mm, kaal: 510 g
  • Ilmumisaeg: 21-Dec-2010
  • Kirjastus: Sage Publications Ltd
  • ISBN-10: 1848601026
  • ISBN-13: 9781848601024
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781848601024.html
Märksõnad:
 



 



Introducing Survival Analysis and Event History Analysis is an accessible, practical and comprehensive guide for researchers and students who want to understand the basics of survival and event history analysis and apply these methods without getting entangled in mathematical and theoretical technicalities. Inside, readers are offered a blueprint for their entire research project from data preparation to model selection and diagnostics.



 



Engaging, easy to read, functional and packed with enlightening examples, hands-on exercises and resources for both students and instructors, Introducing Survival Analysis and Event History Analysis allows researchers to quickly master these advanced statistical techniques. This book is written from the perspective of the user, making it suitable as both a self-learning tool and graduate-level textbook.



 



Introducing Survival Analysis and Event History Analysis covers the most up-to-date innovations in the field, including advancements in the assessment of model fit, frailty and recurrent event models, discrete-time methods, competing and multistate models and sequence analysis. Practical instructions are also included, focusing on the statistical program R and Stata, enabling readers to replicate the examples described in the text.



 



This book comes with a glossary, a range of practical and user-friendly examples, cases and exercises.

Arvustused

This book is very useful for researchers and students in different scientific areas social sciences and humanities, medicine, in



general every science where studies measuring time changes in variables are



conducted...As the author explains, this book is written from the



perspective of an absolute beginner comprehensible and with a lot of examples



in the text, tables and graphs. It goes beyond an introductory textbook on this



topic, because it presents not only non-parametric models, semi-parametric



models, parametric models, model-building and model diagnostics, but it is focused also on some more recent techniques like frailty and recurrent event



history models, discrete-time models, multistate models, competing risk



analysis and sequence analysis...Everyone who would like to start with Survival and



Event History analysis or to get more knowledge of Survival and Event History



analysis could do this by reading this book Stanislava Yordanova Stoyanova Methodspace

List of Tables
xii
List of Figures
xiv
List of Boxes
xvii
Preface xviii
1 The fundamentals of survival and event history analysis
1(17)
1.1 Introduction: what is survival and event history analysis?
1(1)
1.2 Key concepts and terminology
2(2)
1.3 Censoring and truncation
4(3)
1.3.1 Right-censoring
5(1)
1.3.2 Interval censoring
6(1)
1.3.3 Truncation
6(1)
1.4 Mathematical expression and relation of basic statistical functions
7(2)
1.5 Why use survival and event history analysis?
9(2)
1.5.1 Potential problems that might arise if censored data is ignored
9(2)
1.5.2 What does survival analysis offer that ordinary regression models do not?
11(1)
1.6 Overview of survival and event history models and this book
11(7)
1.6.1 Non-, semi- and parametric models
11(3)
1.6.2 Outline of this book
14(3)
Exercises
17(1)
2 An introduction to R and data exploration via descriptive statistics and graphics
18(29)
2.1 An introduction to R and data exploration
18(2)
2.2 Downloading R on your personal computer
20(1)
2.3 The R base system and add-on packages
21(1)
2.3.1 Add-on packages and how to install them
21(1)
2.3.2 Loading an add-on package
22(1)
2.4 Running R
22(3)
2.4.1 Running R interactively by typing at the > prompt
22(1)
2.4.2 Running R non-interactively using a script file
23(1)
2.4.3 Running R using the R Commander graphical user interface
24(1)
2.5 Determining and setting your working directory
25(2)
2.5.1 Determining your working directory
25(1)
2.5.2 Setting a new working directory
26(1)
2.6 Help and documentation
27(1)
2.7 Importing data into R
27(4)
2.7.1 Importing Stata or SPSS data into R
28(2)
2.7.2 Importing ASCII text or Excel data into R
30(1)
2.8 Working with data: opening and accessing variables from a data frame
31(4)
2.8.1 Placing the name of the data within a function
32(1)
2.8.2 Using the $ sign
32(1)
2.8.3 Using (and abusing) the attach function
33(1)
2.8.4 Using data that is part of an existing library package
34(1)
2.8.5 Saving data
34(1)
2.9 Saving your work and quitting R
35(2)
2.9.1 Save to file and capture output options
35(1)
2.9.2 Quitting R and saving your workspace
35(2)
2.9.3 Saving your history
37(1)
2.10 Basic descriptive statistics
37(6)
2.10.1 The example data
37(1)
2.10.2 Descriptive summary statistics
38(5)
2.11 Descriptive data exploration with graphics
43(4)
Exercises
45(2)
3 Survival and event history data structures
47(15)
3.1 Introduction: why discuss data structures?
47(1)
3.2 Sources of event history data
48(1)
3.3 Single-episode data
49(1)
3.4 Multi-episode data
50(3)
3.4.1 Understanding multi-episode data
50(1)
3.4.2 Converting single-episode to multi-episode data
51(2)
3.5 Subject-period (discrete-time) data, episode-splitting and counting process format
53(5)
3.5.1 Subject-period or discrete-time data
53(1)
3.5.2 Creating a subject-period file: survSplit in survival library
54(1)
3.5.3 Creation of a subject-period file: to Binary in eha package
55(1)
3.5.4 Episode-splitting
56(2)
3.5.5 Counting process style of data
58(1)
3.6 A note on dates
58(4)
3.6.1 Using as.Date
59(1)
3.6.2 Converting date variables to a numeric format
59(1)
3.6.3 Using chron
60(1)
Exercises
61(1)
4 Non-parametric methods: the Kaplan-Meier estimator
62(24)
4.1 Introduction
62(1)
4.2 The Kaplan-Meier (KM) estimator
63(1)
4.3 Undertaking KM estimations in R
64(3)
4.3.1 The survival package in R
64(2)
4.3.2 Loading RcmdrPlugin.survival to use in the R Commander
66(1)
4.4 Kaplan-Meier estimation
67(6)
4.4.1 Producing KM estimates using the R Commander
67(2)
4.4.2 Producing KM estimates with a script file
69(2)
4.4.3 Interpretation of KM estimates
71(2)
4.5 Plotting the Kaplan-Meier survival curve
73(6)
4.5.1 Plotting a univariate KM survival curve
73(2)
4.5.2 Comparing two KM survival curves
75(4)
4.6 Testing differences between two groups using survdiff
79(4)
4.6.1 The Fleming-Harrington test
80(1)
4.6.2 The log-rank (Mantel-Haenszel) test
80(1)
4.6.3 The Peto and Peto test
81(1)
4.6.4 Comparing tests: which test to choose?
82(1)
4.7 Stratifying the analysis by a covariate
83(3)
Exercises
85(1)
5 The Cox proportional-hazards regression model
86(28)
5.1 Introduction: The Cox regression model
86(5)
5.1.1 The Cox proportional hazard model with fixed covariates
87(2)
5.1.2 The Cox proportional hazards model with time-varying covariates
89(1)
5.1.3 Why is the Cox model so popular?
90(1)
5.2 Estimating and interpreting the Cox model with fixed covariates
91(9)
5.2.1 The coxph object
91(1)
5.2.2 Estimating the Cox regression model
91(2)
5.2.3 Interpreting covariate estimates in the Cox regression model
93(4)
5.2.4 Significance of the model
97(1)
5.2.5 Plotting the estimated survival function
98(1)
5.2.6 Plotting the estimated survival function by a covariate
99(1)
5.3 The Cox regression model with time-varying covariates
100(14)
5.3.1 Creating a subject-period file to accommodate time-varying covariates
100(3)
5.3.2 Modelling time-varying covariates using person-period data
103(3)
5.3.3 Creating a subject-period file with lagged variables to reduce problems of causal ordering
106(1)
5.3.4 Lagged time-varying covariates to reduce problems of causal ordering
107(1)
5.3.5 Interactions with time as time-dependent covariates: episode-splitting at time intervals
108(5)
Exercises
113(1)
6 Parametric models
114(27)
6.1 Introduction
114(1)
6.2 Relationship of the probability density, hazard and survival function
115(1)
6.3 Proportional hazards (PH) versus accelerated failure time (AFT) models
116(1)
6.4 Specification of parametric models
117(8)
6.4.1 Summary of selected parametric survival distributions
117(1)
6.4.2 The exponential model
118(3)
6.4.3 Piecewise constant exponential model
121(1)
6.4.4 The Weibull model
121(3)
6.4.5 Log-logistic and log-normal models
124(1)
6.4.6 Other parametric models
125(1)
6.5 Estimating parametric survival models using the survival and eha packages
125(1)
6.5.1 Estimating parametric models using the survreg function in the survival library
125(1)
6.5.2 Estimating parametric models using the phreg and aftreg functions in the eha library
126(1)
6.6 Estimation and interpretation of parametric models
126(13)
6.6.1 Exponential model: PH parameterization
126(2)
6.6.2 Exponential model: AFT parameterization
128(5)
6.6.3 Piecewise exponential model: PH and AFT parameterization
133(3)
6.6.4 Weibull model: PH parameterization
136(1)
6.6.5 Weibull model: AFT parameterization
136(1)
6.6.6 Log-logistic and log-normal models: AFT parameterization
137(2)
6.7 What happens if a parametric model is specified incorrectly?
139(2)
Exercises
140(1)
7 Model-building and diagnostics
141(23)
7.1 Introduction
141(1)
7.2 Model-building and selection of covariates and a model
142(4)
7.2.1 Purposeful selection of covariates
142(2)
7.2.2 The decision path to choosing an appropriate model
144(2)
7.3 Assessing the overall goodness of fit of your model
146(3)
7.3.1 The log-likelihood and likelihood ratio tests
146(2)
7.3.2 Akaike information criterion (AIC) and evaluation of standard errors
148(1)
7.4 Testing overall model adequacy: Cox-Snell residuals
149(2)
7.5 Testing the proportional hazards assumption: Schoenfeld residuals
151(6)
7.5.1 Understanding and estimating Schoenfeld residuals
151(3)
7.5.2 Dealing with non-proportional hazards: introducing an interaction effect
154(1)
7.5.3 Dealing with non-proportional hazards: stratifying the data
155(2)
7.6 Checking for influential observations: score residuals
157(3)
7.6.1 What should be done if influential observations are identified?
160(1)
7.7 Assessing nonlinearity: martingale residual and component-plus-residual plots
160(4)
Exercises
163(1)
8 Frailty and recurrent event models
164(15)
8.1 Introduction
164(2)
8.2 Shared frailty: modelling recurrent events and clustering in groups
166(3)
8.2.1 Recurrent events
166(1)
8.2.2 Shared clustering in groups
167(2)
8.3 Additional frailty models: unshared, nested, joint and additive models
169(2)
8.3.1 Individual (unshared) frailty models
169(1)
8.3.2 Nested frailty models
170(1)
8.3.3 Joint and additive frailty models
170(1)
8.4 Estimating frailty models in R
171(1)
8.4.1 Using the frailty function
171(1)
8.4.2 The frailtypack and survrec library in R
171(1)
8.5 Frailty model estimation and interpretation
172(7)
8.5.1 Description of the data
172(1)
8.5.2 Frailty model with a gamma distribution
173(4)
8.5.3 Frailty model with a Gaussian distribution
177(1)
Exercises
178(1)
9 Discrete-time models
179(11)
9.1 Introduction
179(2)
9.2 Discrete-time models
181(3)
9.2.1 Specification of the hazard, survival and cumulative probability density functions
181(1)
9.2.2 Models to estimate discrete-time data: logit, probit and complementary log-log functions
182(2)
9.3 Restructuring data for discrete-time modelling
184(1)
9.4 Estimation and interpretation of discrete-time models
184(5)
9.4.1 Estimation of logit, probit and cloglog discrete-time models
184(3)
9.4.2 Interpretation and comparison of estimates
187(2)
9.5 Advantages and disadvantages of discrete-time models
189(1)
Exercises
189(1)
10 Competing risk and multi-state models
190(23)
10.1 Introduction
190(1)
10.2 Competing risk models
191(4)
10.2.1 Three central techniques to model competing risks
192(1)
10.2.2 The latent or cause-specific approach
192(1)
10.2.3 The cumulative incidence curve (CIC)
193(2)
10.3 Estimating competing risks using the latent versus CIC approach
195(3)
10.3.1 Data preparation and restructuring
195(2)
10.3.2 Estimating CIC estimates and their standard errors
197(1)
10.4 Regression analysis with competing risks
198(4)
10.5 Multi-state models
202(2)
10.5.1 A brief introduction to multi-state models and their applications
202(1)
10.5.2 Markov, semi-Markov and extended Markov model properties
203(1)
10.6 Estimation of multi-state models
204(9)
10.6.1 Preparation of data for multi-state models using the mstate package
204(3)
10.6.2 Estimation of Markov model with stratified hazards
207(3)
10.6.3 Estimation of Markov model with proportional hazards
210(1)
10.6.4 Estimation of state arrival extended Markov proportional hazards model
211(1)
10.6.5 Further predictions and estimation of multi-state models with the cumulative incidence function
212(1)
Exercises
212(1)
11 Sequence analysis
213(14)
11.1 Introduction: sequence analysis
213(2)
11.1.1 A brief introduction to sequence analysis
213(2)
11.1.2 Optimal-matching techniques
215(1)
11.2 Sequence analysis data and estimation using the TraMineR package
215(2)
11.2.1 Sequence data
216(1)
11.2.2 The transition from school to work using the mvad data
216(1)
11.3 Describing and visualizing sequence datasets
217(4)
11.3.1 Exploring the data, sequence frequency and state distribution plots
217(2)
11.3.2 Calculating entropy and turbulence
219(2)
11.4 Measuring similarities and distances between sequences
221(1)
11.5 Producing typologies of trajectories: cluster analysis
221(2)
11.6 Event sequence analysis
223(1)
11.7 Criticisms of the OM approach and the dynamic future of sequence analysis
224(3)
Exercises
225(2)
Appendix 1 Description of the data used in this book 227(5)
Appendix 2 Survival and event history analysis using stata 232(23)
Glossary 255(6)
References 261(12)
Index 273