s
Home | GLMs | Multilevel | Survival | Demography | Stata | R

Biographical Notes

I was trained as a Biostatistician at the University of North Carolina at Chapel Hill, and before that obtained a Master's degree in Social Science at the University of Chicago. My undergraduate work in my native Chile was in Psychology, where I discovered my love for statistics while studying psychometrics with Erika Himmel. A job as a statistical assistant to Anibal Faundes in a fertility survey led to a life-long interest in population.

Following completion of my Ph.D. I worked for seven years at the World Fertility Survey (WFS) in London, England, once described as the largest social science project ever undertaken. I then returned home and spent five years in the Statistics Department at the Universidad Católica de Chile, before coming to the Office of Population Research at Princeton University, where I have been since 1987. I have been lucky to interact with wonderful colleagues in all three institutions.

Welcome to my home page! This website collects a few pages related to my research and teaching. It has six main sections dealing with courses in statistics and demography and with statistical software tutorials. These can always be reached via the links at the very top.

Research

My main research interest is statistical demography, the development and application of statistical modeling techniques to the study of human population. My subject areas include fertility and health.

See the multilevel section for some of my work on multilevel models, including my papers with Noreen Goldman, and links to the datasets used in those papers, including both actual and simulated data.

The Handbook of Multilevel Analysis, edited by Jan de Leeuw and Erik Meijer, was officially published on Friday December 21, 2007. I contributed Chapter 9 on 'Generalized Linear Mixed Models'.

In other recent work I have looked at tempo effects in fertility and mortality, and have a paper in Demographic Research with a critical view of tempo adjustments in mortality estimation. The paper also appeared in a 2008 monograph edited by Barbi, Bongaarts and Vaupel which is available online.

Teaching

Each fall I teach a course on generalized linear models, which covers regression models for continuous data (multiple regression, analysis of variance and analysis of covariance), for binary data (including logistic regression and probit models), for count data (Poisson, over-dispersed Poisson and negative binomial models) and for time to event or survival data (mostly piece-wise exponential hazard models using the Poisson-trick). I used to include a bit on log-linear models for contingency tables, but the last few years I substituted a week on models for longitudinal and panel data, including fixed and random-effects models.

The course website includes a set of lecture notes, available in pdf format as well as html, and a collection of Stata and R logs that show how to obtain practically every result in the notes using the statistical package Stata or the R language for statistical computing and graphics. Depending on the time of the year you also see problem sets and exams, with solutions. Sometimes I hear from students in remote corners of the globe who have found these materials useful.

Every other spring term I also teach half-semester courses, one on Survival Analysis and one on Multilevel Models. These courses were last offered in 2014 and iwll be offered again in 2016. The corresponding sections of the website, while less extensive than the section on GLMs, have a collection of handouts with emphasis on computation.

In the Spring of 2006 I taught Research Methods in Demography. The demography section of the website has handouts that use Stata to do demographic calculations organized under 12 different topics ranging from rates and standardization to stable populations. There are also four problem sets with solutions. I taught this course again in the Spring of 2008, but we put substantially less emphasis on demographic computation and did all calculations in Excel.

Service

In my spare time I have developed the web software used since 2002 by the Population Association of America (PAA) to manage its annual meetings, including online submissions and reviews. The system has also been used by the International Union for the Scientific Study of Population (IUSSP), the European Association for Population Studies (EAPS), and the Union for African Population Studies (UAPS). I am particularly proud of the 2005 IUSSP site, which is available in English, French, and Spanish. All these sites run from the same code base (written in C#), using resource strings for localization.