Foundations of Statistical Analysis in Demography
4 February - 15 February 2013, Moscow, Russia, New Economic School
The course focuses on the statistical estimation of probabilities and incidence of demographic events as well as their timing. It presents methods of classical and advanced Statistics as well as their implementation using the programming language R. All concepts are illustrated by numerous demographic examples.
The course appeals to faculty members, researchers, and students with an interest in demography and related disciplines. Basic knowledge of mathematics, statistics, and demography is preferable, but not mandatory, as all necessary concepts will be diligently and thoroughly introduced. Acquaintance with the course’s programming language R is also desirable, but not compulsory.
1. Agresti A., and B. Finlay (2009). Statistical Methods for the Social Sciences. 4th edition. Pearson Prentice Hall.
2. Rodriguez, G.. Lecture Notes: http://data.princeton.edu/wws509/notes/
3. Klabfleisch, J.D., and R.L. Prentice (1980). The Statistical Analysis of Failure Time Data. New York: Wiley.
I. Introduction: Demographic Data – the Magnitude and Timing of Demographic Events. General Framework of Statistical Analysis.
II. Statistics of Demographic Rates
1. Descriptive Statistics in R: Preliminary Data Analysis in Demography
2. Statistical Models as Data-Generating Mechanisms. Applications of Probability Theory in Demography and Epidemiology
3. Parametric Statistical Models: Maximum Likelihood Estimation
4. Application of Maximum Likelihood Estimation in Demographic Models
a) age-specific mortality rates and childlessness: the Binomial model
b) fecundity (menstrual cycles to conception): the Geometric model
c) death counts: the Poisson model
5. Parametric Statistical Models with Covariates:
a) contraceptive use: Logistic Regression
b) cancer incidence: Poisson Regression
III. Introduction to Statistics of Demographic Durations (Survival Analysis)
1. Special Features of Time-to-Event Data: Censoring and Truncation
2. Examples Parametric Models for Duration Data:
a) non-human mortality: the Weibull model
b) human mortality: the Gompertz model
3. Non-Parametric Models: the Kaplan-Meier Estimate and Cox Regression. Examples.
4. Unobserved Heterogeneity in Human Mortality Data: Frailty Models