 |
| |||||
 |
| |||||||||||||||||||||
Statistical Analysis of Epidemiologic Data, 3rd Edition
Comment from the Stata technical groupStatistical Analysis of Epidemiologic Data provides basic material for a one- or two-term course in statistical methods for epidemiology and is ideal for graduate students in the medical sciences who have had basic courses in statistics and epidemiology. Since the methods covered are directly available in most statistical software, including Stata, the mathematical discussions are minimal. Real-data examples are used to motivate the discussion, allowing the reader to replicate the analyses performed and to compare and contrast different methods applied to the same data. Covered topics include an overview of the relevant terminology (e.g., rates, prevalence, hazard functions, and matched case–control data), variation and bias, power and sample-size calculations for standard univariate analysis, cohort analysis, the analysis of contingency tables, logistic regression, Poisson regression, the analysis of matched data, and survival analysis, including the Cox model. Table of contents1 Measures of Risk: Rates and Probabilities
Rates
Probabilities Incidence and prevalence Survival probabilities and hazard rates Statistical properties of probabilities calculated from mortality or disease data Analysis of rates: smoothing, transforming, and adjusting 2 Variation and Bias
A simple model
The t-test
Selection bias Confounder bias Ecologic bias Comparison of k groups Interaction contrasts Two-way analysis Misclassification bias 3 Statistical Power and Sample Size Calculations
Technical details
Normal distribution: sample mean Poisson distribution: relative risk Sample size: one-sample test of a proportion Sample size: two-sample test of proportions Loss of statistical power and bias from grouping continuous data 4 Cohort Data: Description and Illustration
Cohort effect: model
Birth cohort effect and proportional mortality data Median polish analysis Mean polish analysis Four examples Cohort effect: prostatic cancer data 5 Spatial Data: Analysis and Estimation
Poisson model
Nearest-neighbor analysis Transformed maps Spatial distribution about a point Time/distance spatial analysis Randomization test Time/distance: randomization test Bootstrap estimation and analysis 6 The 2 x k Table and the 2 x 2 x 2 Table
The 2 x k contingency table
Independence and homogeneity Regression Two-sample model: comparison of two means Ridit probability analysis The 2 x 2 x 2 contingency table 7 The Analysis of Contingency Table Data: Logistic Model I
The simplest model: discrete case
The 2 x 2 x 2 table The 2 x k table The 2 x 2 x k table The multiway table Goodness-of-fit: discrete case Summarizing a series of 2 x 2 tables 8 The Analysis of Binary Data: Logistic Model II
Simple logistic regression: continuous cas
e Bivariate logistic regression Logistic regression coefficients: general considerations "Centering" The WCGS additive logistic model Case/control sampling 9 The Analysis of Count Data: Poisson Model
Simplest Poisson model
Poisson model: technical description Illustration of the Poisson model Poisson model adjusted rate: Hodgkins disease Poisson regression: CHD risk by smoking and behavior type Count data: application of the Poisson model Poisson regression: a two-way classification Poisson regression: a three-way classification 10 The Analysis of Matched Data: Three Approaches
Frequency matching
Poststratification Matched analysis: continuous variable Matched case/control pairs: binary risk factor Confidence interval for the odds ratio Evaluating the estimated odds ratio Disregarding matching Interaction with the matching variable Matched sets using more than one control Matched analysis: multilevel categorical risk factor Conditional logistic analysis 11 Life Table Analysis: An Introduction
Complete, current life table: construction
Life table survival function Life table hazard function Life tables: three applications of life table techniques Competing Risks 12 Survival Data: Estimation of Risk
Parametric model
Age adjustment of rates Censored and truncated data Mean survival time: parametric estimate Mean survival time: nonparametric estimate Mean survival time from censored data Goodness-of-fit Two-sample data The Wilcoxon test and the Gehan generalization to survival data 13 Survival Data: Proportional Hazards Model
Simplest case
The proportional hazards model Plotting survival curves Four applications of a proportional hazards model Dependence on follow-up time Appendix
A. Description of the WCGS data set
B. Binomial and Poisson probability distributions C. The odds ratio and its properties D. Partitioning the chi-square statistic E. Maximum likelihood estimation and likelihood functions F. Problems References
Index
|
|||||||||||||||||||||
){kind=link}
){kind=link}