Hi all,
Unfortunately, today's seminar is canceled. Instead, we will have an
informal session on the regular zoom link:
https://harvard.zoom.us/j/99424949004?pwd=aWtPNFM3ZzFYbWxIMXNoZDlyUElVZz09
Best,
Soichiro
---------- Forwarded message ---------
From: Thach, Elizabeth <elizabeth_thach(a)fas.harvard.edu>
Date: Wed, Oct 28, 2020 at 10:37 AM
Subject: Cancellation: Opportunity Insights Lunch and IQSS Joint talk
Wednesday, October 28th at 12 noon (EST)
To:
Good morning,
Unfortunately, due to unavoidable circumstances, today’s join lunch session
with the IQSS is cancelled. Please join us for the next OI lunch on
*November
11*, when Douglas Staiger (Dartmouth College), Sean Reardon (Stanford
University), and Thomas J. Kane (Harvard) are presenting.
Please see the attached revised schedule which reflects all changes to the
semester’s schedule, namely that there is no lecture on Nov. 4. Rediet
Abebe will be speaking in the spring. Please let me know if you have any
questions.
Best,
Liz
Elizabeth Thach
Executive Assistant
(p) 617-496-5409
elizabeth_thach(a)fas.harvard.edu
*My pronouns are: she, her, hers*
[image: tiny logo]
Hi all,
Our next virtual meeting will be Wednesday, October 21, where we will hear
Eric Tchetgen Tchetgen presents research on "An Introduction to Proximal
Causal Learning".
*Abstract*: A standard assumption for causal inference from observational
data is that one has measured a sufficiently rich set of covariates to
ensure that within covariates strata, subjects are exchangeable across
observed treatment values. Skepticism about the exchangeability assumption
in observational studies is often warranted because it hinges on one's
ability to accurately measure covariates capturing all potential sources of
confounding. Realistically, confounding mechanisms can rarely if ever, be
learned with certainty from measured covariates. One can therefore only
ever hope that covariate measurements are at best proxies of true
underlying confounding mechanisms operating in an observational study, thus
invalidating causal claims made on basis of standard exchangeability
conditions. Causal learning from proxies is a challenging inverse problem
which has to date remained unresolved. In this paper, we introduce a formal
potential outcome framework for proximal causal learning, which while
explicitly acknowledging covariate measurements as imperfect proxies of
confounding mechanisms, offers an opportunity to learn about causal effects
in settings where exchangeability on the basis of measured covariates
fails. Sufficient conditions for nonparametric identification are given,
leading to the proximal g-formula and corresponding proximal g-computation
algorithm for estimation, both generalizations of Robins' foundational
g-formula and g-computation algorithm, which account explicitly for bias
due to unmeasured confounding. Both point treatment and time-varying
treatment settings are considered, and an application of proximal
g-computation of causal effects is given for illustration.
*Link to paper*: https://arxiv.org/abs/2009.10982
*Zoom link*:
https://harvard.zoom.us/j/99424949004?pwd=aWtPNFM3ZzFYbWxIMXNoZDlyUElVZz09
*Schedule of the workshop*:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Best,
Soichiro
--
Soichiro Yamauchi
PhD candidate
Harvard University
URL: https://soichiroy.github.io/
Hi all,
Our next virtual meeting will be Wednesday, October 14, where we will hear
Luke Miratrix (Harvard University) presents research on "A Practitioner’s
Guide to Intent-to-Treat Effects from Multisite (blocked) Individually
Randomized Trials: Estimands, Estimators, and Estimates".
*Abstract*:
There are many ways to estimate an overall average effect of a large-scale
multisite individually randomized control trial. The researcher can target
the average effect across individuals or sites. Furthermore, the researcher
can target the effect for the experimental sample or a larger population.
If treatment effects vary across sites, these estimands can differ. Once an
estimand is selected, an estimator must be chosen. Standard estimators,
such as fixed-effects regression, can be biased. We describe 15 different
estimators commonly in use, consider which estimands they are appropriate
for, and discuss their properties in the face of cross-site effect
heterogeneity. Using data from 12 large multisite RCTs, we estimate the
effect (and standard error) using each estimator and compare the results.
We assess the extent that these decisions matter in practice and provide
guidance for applied researchers.
Zoom link:
https://harvard.zoom.us/j/99424949004?pwd=aWtPNFM3ZzFYbWxIMXNoZDlyUElVZz09
Schedule of the workshop:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Best,
Soichiro
--
Soichiro Yamauchi
PhD candidate
Harvard University
URL: https://soichiroy.github.io/
Hi all,
Our next virtual meeting will be Wednesday, October 7, where we will hear
Felix Elwer (University of Wisconsin-Madison) presents research on
"Neighborhood Effects in Time." This talk is a joint session with
Opportunity Insights and the talk will not be recorded.
A* different zoom link* will be used for this week:
https://harvard.zoom.us/j/92029171055?pwd=VzZLTUgzWGU3Skl0Wm9yRUVPcGNjUT09
Password: economics
Best,
Soichiro Yamauchi
--
Soichiro Yamauchi
PhD candidate
Harvard University
URL: https://soichiroy.github.io/