Each semester MDI sponsors technical workshops to provide faculty, students, and staff the opportunity to be exposed to new methods, programming paradigms, and technologies. These workshops are led by MDI postdoctoral fellows.
- Please note that at this time, MDI Workshops are only open to Georgetown University students, faculty and staff. Please use your Georgetown email address to register.
- Each workshop takes place over two days from 4:00-5:30pm of each month (January / February / March / April). Attendees should plan to attend both dates.
Please RSVP for each workshop below. For more information on MDI events, check out our events calendar. Or you can add MDI events to your Google calendar.
January 18-19, 2023 – Modeling Propensity to Use or Sell Drugs at the Block Group Level with Dr. J.J. Naddeo
We will use publicly available survey data to build machine learning models to predict the probability an individual recently sold or consumed illicit drugs. We will then connect this model to publicly available census data to investigate patterns in drug use and sales at the public use microdata area (PUMA) level. Finally, utilizing proprietary data from RTI International, we will create maps of drug behavior. We will see how these tools can be used to inform policy makers about who, and where, drugs are being used and sold. Students will code examples using Python. Basic familiarity is expected.
(Previously scheduled for November 2022.)
February 15-16, 2023 – Monte Carlo Simulation and Resampling Methods with Dr. Le Bao
Simulation and resampling methods have been an essential set of contemporary statistical and computational techniques. The underlying logic of using randomness to solve problems also provides a more realistic understanding of the probabilistic interpretation of real-world events. In this workshop, we will cover a wide range of topics related to Monte Carlo simulation, from basic ideas of Monte Carlo methods, to reject sampling, simulating statistical models, and resampling techniques such as bootstrapping and cross validation, and implement them in programming languages. Some prior experience with basic programming and statistical methods will be expected.
March 14-15, 2023– Bayesian Simulation with Dr. Nathan Wycoff
Modern science is characterized by careful quantification of uncertainty. The Bayesian formalism is a widely applicable means of propagating uncertainty in general models. In this workshop, we will begin with a discussion of the basics of Bayesian statistics, before embarking on a program covering Bayesian simulation via Markov-chain Monte Carlo. We will then proceed to various applications related to the social sciences. The goal of this workshop is to give participants basic competency in Bayesian concepts and in software for probabilistic programming. Some prior experience with basic programming and probability will be expected.
April 18-19, 2023 – Causal Inference with Dr. J.J. Naddeo
Everyone has probably heard that correlation does not equal causation. Said another way, just because X and Y are observed to move together it does not mean that X causes Y. In fact, Y may cause X! When working with observational data (i.e. data not generated from an experiment) it is often difficult to know when one can claim X causes Y. The goal of this workshop is to first show how this problem operates through simple simulations. We will then go over two of the most important methods for solving these issues, namely difference in differences and instrumental variables.
Any questions, please reach out to firstname.lastname@example.org
About the Massive Data Institute (MDI): At Georgetown’s McCourt School of Public Policy, the Massive Data Institute (MDI) is an interdisciplinary research institute that connects experts across computer science, data science, public health, public policy, and social science to tackle societal scale issues and impact public policy in a way that improves people’s lives through responsible evidence-based research. For more information on MDI, please visit https://mdi.georgetown.edu/