Adrian Raftery’s course in Paris

UW Professor and U.S. National Academy of Sciences member Adrian Raftery has received the 2020 FSMP research chair and as a result will be visiting Paris this Fall 2021. He will be located at the MAP5 laboratory at the University of Paris. In particular, he will give a 20-hour Master course on statistical semography. This will be given over four successive Tuesdays, with 5 hours of lectures per week. The course is open to all. Attendance is free of charge but registration is mandatory. (To register, please fill the attached form. Lectures will be given in the salle du conseil, on the 7ft floor of the Saint-Germain-des-Prés campus, 45 rue des Saints-Pères, 75006 Paris.)

Demography aims to estimate and forecast population, fertility, mortality and migration. This is important for government policy-making, private sector planning, and research in the health and social sciences, and also critical for climate science and global health. It has traditionally been done using deterministic mathematical methods, but these ignore uncertainty and measurement error.  In the past decade, modern statistical methods were developed for this by our group at the University of Washington, and these were recently adopted by the  United Nations for their official population forecasts for all countries.  Statistical demography is expanding rapidly,  and this course will teach theory and practice of  methods and models of the field, with a focus on current and potential future research.

The topics will be:

1. Review of basic mathematical demographic methods.

2. Modeling age-specific rates, including model schedules and Lee-Carter method.

3. Statistical modeling and projection of fertility, mortality, migration and population.

4. Reconstructing population and vital rates from imperfect data.

Methow River fire

deadend of the day [no filter]

a Bayesian interpretation of FDRs?

This week, I happened to re-read John Storey’ 2003 “The positive discovery rate: a Bayesian interpretation and the q-value”, because I wanted to check a connection with our testing by mixture [still in limbo] paper. I however failed to find what I was looking for because I could not find any Bayesian flavour in the paper apart from an FRD expressed as a “posterior probability” of the null, in the sense that the setting was one of opposing two simple hypotheses. When there is an unknown parameter common to the multiple hypotheses being tested, a prior distribution on the parameter makes these multiple hypotheses connected. What makes the connection puzzling is the assumption that the observed statistics defining the significance region are independent (Theorem 1). And it seems to depend on the choice of the significance region, which should be induced by the Bayesian modelling, not the opposite. (This alternative explanation does not help either, maybe because it is on baseball… Or maybe because the sentence “If a player’s [posterior mean] is above .3, it’s more likely than not that their true average is as well” does not seem to appear naturally from a Bayesian formulation.) [Disclaimer: I am not hinting at anything wrong or objectionable in Storey’s paper, just being puzzled by the Bayesian tag!]

photography exhibit in Seattle

For readers in the Seattle vicinity, just mentioning that a friend and climbing guide, Brittany Aäe, is holding a photograph exhibit at Caffe Vita, 4301 Fremont Avenue North, in Seattle, this month. Enjoy.