Dear Applied Statistics Workshop Community,
Just a quick reminder, our next meeting is Wednesday, April 3 (12:00 EST).
Zeyang Yu will present "A Binary IV Model for Persuasion: Profiling
Persuasion Types among Compliers."
<When>
April 3, 12:00 to 1:30 PM, EST
Lunch will be available for pick-up inside CGIS K354.
<Where>
In-person: CGIS K354
Zoom:
https://harvard.zoom.us/j/93217566507?pwd=elBwYjRJcWhlVE5teE1VNDZoUXdjQT09
<Abstract>
In the empirical study of persuasion, researchers often use a binary
instrument to encourage individuals to consume information and take some
action. We show that with the Imbens-Angrist instrumental variable model
assumptions and the monotone treatment response assumption, it is possible
to identify the joint distributions of potential outcomes among compliers.
This is necessary to identify the percentage of persuaded individuals and
their statistical characteristics. Specifically, we develop a weighting
method that helps researchers identify the statistical characteristics of
persuasion types: compliers and always-persuaded, compliers and persuaded,
and compliers and never-persuaded. These findings extend the ”κ weighting”
results in Abadie (2003). We also provide a sharp test on the two sets of
identification assumptions. The test boils down to testing whether there
exists a nonnegative solution to a possibly under-determined system of
linear equations with known coefficients. An application based on Green et
al. (2003) is provided. The result shows that among compliers, roughly 10%
voters are persuaded. The results are consistent with the findings that
voters’ voting behaviors are highly persistent.
Link to the paper: yu_2023local.pdf
<https://arthurzeyangyu.github.io/jmp/yu_2023local.pdf>
<2023-2024 Schedule>
GOV 3009 Website:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Calendar:
https://calendar.google.com/calendar/u/0?cid=Y18zdjkzcGF2OWZqa2tsZHJidTlzbm…
Best,
Jialu
--
Jialu Li
Department of Government
Harvard University
*https://jialul.github.io/ <https://jialul.github.io/>*
Dear Applied Statistics Workshop Community,
Our next meeting will be on Wednesday, March 27 (12:00 EST). Shuangning Li
presents "Experimenting under Stochastic Congestion."
<When>
March 27, 12:00 to 1:30 PM, EST
Lunch will be available for pick-up inside CGIS K354.
<Where>
In-person: CGIS K354
Zoom:
https://harvard.zoom.us/j/93217566507?pwd=elBwYjRJcWhlVE5teE1VNDZoUXdjQT09
<Abstract>
We study randomized experiments in a service system when stochastic
congestion can arise from temporarily limited supply and/or demand. Such
congestion gives rise to cross-unit interference between the waiting
customers, and analytic strategies that do not account for this
interference may be biased. In current practice, one of the most widely
used ways to address stochastic congestion is to use switchback experiments
that alternatively turn a target intervention on and off for the whole
system. We find, however, that under a queueing model for stochastic
congestion, the standard way of analyzing switchbacks is inefficient, and
that estimators that leverage the queueing model can be materially more
accurate. We also consider a new class of experimental design, which can be
used to estimate a policy gradient of the dynamic system using only
unit-level randomization, thus alleviating key practical challenges that
arise in running a switchback.
<2023-2024 Schedule>
GOV 3009 Website:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Calendar:
https://calendar.google.com/calendar/u/0?cid=Y18zdjkzcGF2OWZqa2tsZHJidTlzbm…
Best,
Jialu
--
Jialu Li
Department of Government
Harvard University
*https://jialul.github.io/ <https://jialul.github.io/>*
Dear Applied Statistics Workshop Community,
Our next meeting will be on Wednesday, March 20 (12:00 EST). Anton
Strezhnev presents "A Guide to Dynamic Difference-in-Differences
Regressions for Political Scientists."
<When>
March 20, 12:00 to 1:30 PM, EST
Lunch will be available for pick-up inside CGIS K354.
<Where>
In-person: CGIS K354
Zoom:
https://harvard.zoom.us/j/93217566507?pwd=elBwYjRJcWhlVE5teE1VNDZoUXdjQT09
<Abstract>
Difference-in-differences (DiD) designs for estimating causal effects have
grown in popularity throughout political science. Many DiD papers present
their central results through an "event study" plot - a visualization that
combines estimated dynamic average treatment effects for multiple
post-treatment time periods alongside placebo tests of the main identifying
assumption: parallel trends. Despite their ubiquity, the methods used in
practice for the creation of these plots are not standardized and in many
cases the approaches adopted by researchers can result in misleading
inferences about both the treatment effects and the placebo tests. Building
on and synthesizing recent contributions in the econometric literature on
differences-in-differences designs, this paper illustrates some common
pitfalls through a replication of three recently published papers in major
political science journals. We identify three notable problems related to
the incorrect specification of the baseline comparison time, incorrect
inclusion of "always-treated" units, and sensitivity to effect homogeneity
assumptions. We help provide researchers with additional intuition for the
problems that arise due to effect heterogeneity and for the "contamination
bias" result of Sun and Abraham (2021) through a novel decomposition of the
dynamic event study regression in the style of Goodman-Bacon (2021) that
allows researchers to recover the weights placed on each 2x2 comparison
used to construct the effect estimates and placebos. These weights allow
researchers to gauge the sensitivity of each estimate to potential effect
heterogeneity.
Anton is happy to meet with students and faculty after the talk. Please
reach out to Jialu directly if you want to schedule 1:1 meetings with him.
<2023-2024 Schedule>
GOV 3009 Website:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Calendar:
https://calendar.google.com/calendar/u/0?cid=Y18zdjkzcGF2OWZqa2tsZHJidTlzbm…
Best,
Jialu
--
Jialu Li
Department of Government
Harvard University
*https://jialul.github.io/ <https://jialul.github.io/>*
Dear Applied Statistics Workshop Community,
Our next meeting will be on Wednesday, March 6 (12:00 EST). Amanda
Coston presents "Addressing confounding in decision-making algorithms."
<When>
March 6, 12:00 to 1:30 PM, EST
Lunch will be available for pick-up inside CGIS K354.
<Where>
In-person: CGIS K354
Zoom:
https://harvard.zoom.us/j/93217566507?pwd=elBwYjRJcWhlVE5teE1VNDZoUXdjQT09
<Abstract>
Machine learning algorithms are used for decision-making in societally
high-stakes settings from child welfare and criminal justice to healthcare
and consumer lending. These algorithms are often intended to predict
outcomes under a proposed decision. It is challenging to evaluate how well
these algorithms perform because we only observe the relevant outcome under
a biased sample of the population. In this talk, we explore how to use
techniques from causal inference to estimate performance on the full
population. We will consider several strategies to account for confounding
factors that affect the decision and the outcome. First, we study runtime
confounding where all relevant factors are captured in the historical data,
but it is either undesirable or impermissible to use some such factors in
the prediction model. Second, we study the setting with unobserved
confounders where we can bound the degree to which the outcome varies on
average between units receiving different decisions conditional on observed
covariates and identified nuisance parameters. We develop debiased machine
learning estimators for the learning target and predictive performance
estimands under both settings. We present empirical results in the consumer
lending and child welfare domains.
Papers: arxiv:2212.09844 <https://arxiv.org/abs/2212.09844> and
arxiv:2006.16916 <https://arxiv.org/abs/2006.16916>.
<2023-2024 Schedule>
GOV 3009 Website:
https://projects.iq.harvard.edu/applied.stats.workshop-gov3009
Calendar:
https://calendar.google.com/calendar/u/0?cid=Y18zdjkzcGF2OWZqa2tsZHJidTlzbm…
Best,
Jialu
--
Jialu Li
Department of Government
Harvard University
*https://jialul.github.io/ <https://jialul.github.io/>*