Econometrics postdoc and computational statistics postdoc openings here in the Stan group at Columbia

Andrew and I are looking to hire two postdocs to join the Stan group at Columbia starting January 2020. I want to emphasize that these are postdoc positions, not programmer positions. So while each position has a practical focus, our broader goal is to carry out high-impact, practical research that pushes the frontier of what’s posisble in Bayesian modeling. This particular project is focused on extremely challenging econometric modeling problems and statistical computation and will be carried out in conjunction with some really great economists (details in the job descriptions below).

These positions are funded through a generous gift from the Alfred P. Sloan Foundation.

Computational statistics postdoc

The Stan group at Columbia is looking to hire a Postdoctoral Research Scholar to work on computational statistics. The goal of the project is to:

* develop algorithms for solving differential and algebraic equations, potentially stochastic and partial

* fit large scale-hierarchical models either through core sampling improvements or approximations such as nested Laplace or variational inference.

In both projects, there is wide latitude for extending the state of the art in computational statistics. The Stan project encompasses a team of dozens of developers distributed around the world and this work will be done in collaboration with that wider team. The wider team provides expertise in everything from numerical analysis and applied mathematics to programming language theory and parallel computation. The position is well funded to travel to conferences and visit collaborators.

The project is funded through a grant focused on Bayesian econometric modeling, which provides concrete applications that will provide a focus for the work as well as a second postdoc funded to develop those applications concurrently with developing the tools needed to extend the existing state of the art. The Stan group at Columbia is also working on applications of differential and algebraic equations in soil carbon modeling and pharmacology and applications of large scale hierarchical models in education and in survey sampling for political science.

The position will be housed in the Applied Statistics Center at Columbia University and supervised by Bob Carpenter. The initial appointment will be for 18 months (January 2020 through June 2022) with a possibility of extension.

Columbia is an EEO/AA employer

To apply, please send a CV and a statement of interest and experience in this area if not included in the CV to Bob Carpenter, [email protected]. The position is available starting in January 2020, and we will review applications as they arrive.

Econometrics Postdoc

The Stan group at Columbia is looking to hire a Postdoctoral Research Scholar to work on Bayesian econometric modeling and methodology. The goal is to create a bridge from modern econometric modeling to current Bayesian computational practice by generating a range of illustrative case studies illustrating best practices. Many of these best practices will need to be developed from the ground up and there is wide latitude for novel work.

This work will be carried out in collaboration with several economists and methodologists outside of Columbia University:

* Empirical auction analysis, where the theory around optimal design can be used to improve econometric methods used to draw inferences from the performance of real auctions in practice, including jointly modeling all components of a bidding system in order to test the structural assumptions driving mechanism design decisions. With Prof. Shoshanna Vasserman (Stanford)

* Bounded rationality and decision making in dynamic and stochastic environments, where macroeconomic models may be expressed in the form of dynamic, stochastic, general equilibrium models which can be extended to higher orders to model bounded rationality in agents making decisions in dynamic and stochastic environments. With Prof. Thomas Sargent, New York University.

* External validity of policy targeting for subgroups, with the goal of applying interventions where they will benefit the most while avoiding harming other subgroups, and a focus on combining data across multiple settings using meta-analysis. With Prof. Rachel Meager, London School of Economics.

* Causal models of interventions in education policy, where the focus is on time-series data organized by classroom, school, and larger groupings in the context of heterogeneous demographic. With Prof. Sophia Rabe-Hesketh, University of California, Berkeley.

Basic capabilities to fit these models exist in Stan currently and this grant will support a second postdoc to help extend those capabilities to more complicated systems.

The position will be housed in the Applied Statistics Center at Columbia University and supervised by Andrew Gelman. The initial appointment will be for 18 months (January 2020 through June 2022) with a possibility of extension.

Columbia is an EEO/AA employer

To apply, please send a CV and a statement of interest and experience in this area if not included in the CV to Andrew Gelman, at [email protected]. The position is available starting January 1, 2020, and we will review applications as soon as they arrive.

Another day, another stats postdoc

This post is from Phil Price.  I work in the Environmental Energy Technologies Division at Lawrence Berkeley National Laboratory, and I am looking for a postdoc who knows substantially more than I do about time-series modeling; in practice this probably means someone whose dissertation work involved that sort of thing.  The work involves developing models to predict and/or forecast the time-dependent energy use in buildings, given historical data and some covariates such as outdoor temperature.  Simple regression approaches (e.g. using time-of-week indicator variables, plus outdoor temperature) work fine for a lot of things, but we still have a variety of problems.  To give one example, sometimes building behavior changes — due to retrofits, or a change in occupant behavior — so that a single model won’t fit well over a long time period. We want to recognize these changes automatically .  We have many other issues besides: heteroskedasticity, need for good uncertainty estimates, ability to partially pool information from different buildings, and so on.  Some knowledge of engineering, physics, or related fields would be a plus, but really I just need someone who knows about ARIMA and ARCH and all that jazz and is willing to learn the rest. If you’re interested, apply through the LBNL website.