Mixtape Sessions: Year Two
Spring 2023 schedule is up!
Introducing Mixtape sessions: Year Two
One of my favorite comic books is Frank Miller’s Batman: Year One. It tells the story of Bruce Wayne learning the ropes the first year he wore the cowl in Gotham City. It’s dark, moody, violent, filled with close calls and excitement as Wayne gradually learns who he is, who his enemies are, and who his friends are. That sounds foreboding, but it’s my way of simply saying that Mixtape Sessions: Year One has concluded and it was a great year. Between the Mixtape Sessions platform and the more specialized workshops for agencies, departments and firms, around 2600 people attended one of our workshops taught by either myself, Peter Hull, Jon Roth, Mark Anderson and Dan Rees or Brigham Frandsen. It was a great experiment to be part of, and I am excited therefore to announce there will be a sequel — Mixtape Sessions: Year Two. You can find the updated list of workshops here at the website, but in case you are wanting to also see it yourself, then let me briefly give you a walk through.
Causal Inference Flagship Workshops
There are two “flagship” workshops at Mixtape Sessions, both taught by me. They more or less follow the outline of second edition of my book, Causal Inference: the Mixtape, which I’m busy writing (mid-March deadline!). There’s a 4-day, 8-hours a day, Causal Inference Part I, and a 4-day, 8-hours a day, Causal Inference Part II. Participants get access to recordings posted to Vimeo (password protected) as well as permanent access to the Discord channel that we use for chat. The interface is over Zoom.
Please be sure to tell anyone you know about this if you think they’d like it or benefit from it. Our goal is to use price discrimination aggressively to help those with low ability to pay but high need to be able to take these courses. And even for those paying full price, you can shop around and see that these prices are usually lower than alternatives out there. We hope that this can be a useful experience for you. You can email me at causalinf@mixtape.consulting.
Now let me tell you a little bit about the courses now.
1) Causal Inference Part I (taught by Scott Cunningham at Baylor University). This 4-day workshop covers the foundations of contemporary causal inference by starting with the Neyman-Rubin potential outcomes model and Pearl’s causal graphs. It discusses the role that randomization places in estimating causal effects, problems of selection bias, and SUTVA before going into three dominant research designs: selection on observables, regression discontinuity design and instrumental variables. We review the theory behind each, a variety of estimators associated with them, detailed examples and applications, and walk-through “lab exercises” in Stata and R. This is the prequel to a Part II course called Causal Inference Part II that covers difference-in-differences and synthetic control.
February 4-5 and 11-12
8:00am to 5:00pm CST (15 min breaks every hour and a 1-hour lunch break at noon)
https://www.mixtapesessions.io/session/ci_i_feb4/
New: Limited $1 spots for residents of low income countries. (contact for promo code to see if you quality)
$50 for students, predocs, postdocs and residents of middle income countries. See website for schedule. (contact for promo code)
$595 for all others
2) Causal Inference Part II (taught by Scott Cunningham at Baylor University). This 4-day workshop is the sequel to Causal Inference Part I. It covers two contemporary research designs that have over the last few years grown in importance and changed a great deal: difference-in-differences and synthetic control. This material will cover the basics of diff-in-diff, as well as more complex issues such as event studies, triple differences, the inclusion of covariates, and differential timing. If we have time, we will also cover continuous treatments. Synthetic control will cover two methods: the canonical model by Alberto Abadie (Abadie and Gardezeabal 2003; Abadie, Diamond and Hainmueller 2010) and augmented synthetic control (Ben-Michael, Feller and Rothstein 2021). We will review the theory behind each design, go into detail on the intuition of the estimation strategies as well as the explicit calculations, and explore code in R and Stata using coding labs. The goal is for you to leave the workshops with competency, confidence and comprehension.
March 18-19 and 25-26
8:00am to 5:00pm CST (15 min breaks every hour and a 1-hour lunch break at noon)
https://www.mixtapesessions.io/session/ci_ii_mar18/
New: Limited $1 spots for residents of low income countries. (contact for promo code to see if you quality)
$50 for students, predocs, postdocs and residents of middle income countries. See website for schedule. (contact for promo code)
$595 for all others
Mixtape Tracks
In addition to our flagship workshops on causal inference (Causal Inference I and II), we offer shorter workshops by leading experts that we call “Mixtape Tracks”. These are typically 6-8 hours spread over 1-2 days. Same rules apply — you get access to video recordings uploaded to Vimeo (password protected) and the Discord channel used for chatting. These are chances to go deeper on important select topics in applied econometrics from some of the top econometricians in the world.
Machine Learning and Causal Inference
3) Machine Learning and Causal Inference (taught by Brigham Frandsen), February 23-24th. Machine Learning’s wheelhouse is out-of-sample prediction, but these powerful methods can be deployed in service of causal inference. This two-session workshop will introduce the basics of machine learning prediction methods, including lasso and random forests and how they feature in causal inference methods like double machine learning and post-double selection lasso. The course covers the conceptual and theoretical basis for the methods and also gets into the nuts and bolts of implementation in python and Stata using real-world data.
Starting February 23-24
6:00pm to 9:00pm EST
https://www.mixtapesessions.io/session/ml_oct28/
New: Limited $1 spots for residents of low income countries. (contact for promo code to see if you quality)
$50 for students, predocs, postdocs and residents of middle income countries. See website for schedule. (contact for promo code)
$595 for all others
4) Machine Learning and Causal Inference (taught by Brigham Frandsen), May 15th. The holy grail of causal inference is the individual-level treatment effect: how would a particular patient respond to a drug? Which users will respond most to a targeted ad? Would a given student be helped or harmed by a classroom intervention?This session introduces machine learning tools for estimating heterogenous treatment effects like random causal forests. The course goes over the theory and concepts as well as the nitty-gritty of coding the methods up in python, R and Stata using real-world examples. This course can be taken as a follow-up to the Machine Learning and Causal Inference Mixtape Session (also taught by Brigham) or as a stand-alone course.
May 15th
6:00pm to 9:00pm EST
https://www.mixtapesessions.io/session/ml_het_effects_may15/
New: Limited $1 spots for residents of low income countries. (contact for promo code to see if you quality)
$50 for students, predocs, postdocs and residents of middle income countries. See website for schedule. (contact for promo code)
SPECIAL PRICE: $295 for all others
Synthetic Control and Clustering Track
5) Synthetic control and clustering (taught by Alberto Abadie at MIT). Description coming
April 27th and April 28th
Times TBD (set for evening eastern time)
https://www.mixtapesessions.io/session/synth_and_clust_apr27/
New: Limited $1 spots for residents of low income countries. (contact for promo code to see if you quality)
$50 for students, predocs, postdocs and residents of middle income countries. See website for schedule. (contact for promo code)
$595 for all others
Instrumental Variables Tracks
6) Introducing Instrumental Variables (taught by Peter Hull at Brown University). This two day workshop is a focused introduction to instrumental variables, one of the most powerful methods of detecting causal effects with observational data or non-compliance with randomized experiments. Peter covers such topics as two stage least squares, marginal treatment effects, various issues created by heterogeneity and the judge leniency design, and includes coding labs.COMING SOON IN 20237) Shift-Share (“Bartik IV”) Instrumental Variables (taught by Peter Hull from Brown University). A popular application of instrumental variables within trade, labor and urban economics is Bartik IV, or “shift-share”. This workshop will introduce the participant to the basics of shift-share IV and cover recent work on the subject in applied econometrics (some of which Peter has written). Group programming exercsies will be used to help illustrate various theoretical concepts in real world applications.COMING SOON IN 2023Hidden Curriculum
In addition to our flagship workshops on causal inference (Causal Inference I and II) and our Mixtape Tracks, we are also offering short workshops that help prepare graduate students and junior faculty on the “hidden curriculum” of moving through the final part of the dissertation stage and through the path to tenure after getting your tenure track job.
Doing Applied Causal Research
8) Doing Applied Research (taught by Mark Anderson and Dan Rees Track), 2-day workshop (8 hours total). Causal inference is one part theory and one part application. We often learn the theory, but we are often left to our own to learn the application. This workshop is intended to be a practical guide for graduate students and early career economists doing applied research. The nuts and bolts of writing, publishing, and service to the profession are covered in two half-day sessions, each lasting roughly four hours (including short breaks). We begin by providing tips on how to start a research project, when to switch topics, and how to effectively manage multiple projects at once. Next, we provide practical advice on how to write an applied economics paper, from structuring the introduction to crafting the conclusion. The second half of the course takes participants through the publication process. In addition, we discuss networking, refereeing for economics journals, getting the most out of conferences, and how to successfully navigate the academic job market. You can find the link here:
May 4-5th
12:00pm to 4:30pm EST
https://www.mixtapesessions.io/session/applied_may4/
The doing applied research workshop will use uniform pricing of $125 per person.
Please be sure to tell anyone you know about this if you think they’d like it or benefit from it. Our goal is to use price discrimination aggressively to help those with low ability to pay but high need to be able to take these courses. And even for those paying full price, you can shop around and see that these prices are usually lower than alternatives out there. We hope that this can be a useful experience for you. You can email me, again, at causalinf@mixtape.consulting
Sincerely,
Scott Cunningham