Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming *Edo
Airoldi*, Professor of Statistics at Harvard University. He will be
presenting joint work with Don Rubin and Daniel Sussman entitled *Estimating
Causal Effects in the Presence of Interfering Units**.* Please find the
abstract below and on the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/presentations…>
.
As usual, we will meet in CGIS Knafel Room 354 from noon to 1:30pm, and
lunch will be provided. See you all there! To view previous Applied
Statistics presentations, please visit the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/videos>.
-- Aaron Kaufman
Title: Estimating Causal Effects in the Presence of Interfering Units
Abstract: Classical approaches to causal inference largely rely on the
assumption of “lack of interference”, according to which the outcome of an
individual does not depend on the treatment assigned to others. In many
applications, however, such as designing and evaluating the effectiveness
of healthcare interventions that leverage social structure, assuming lack
of interference is untenable. In fact, the effect of interference itself is
often an inferential target of interest. In this talk, we will discuss
technical issues that arise in estimating causal effects when interference
can be attributed to a network among the units of analysis, and develop a
strategy for optimal experimental design in this context that involves a
piecewise constant approximation of a certain graphon.
--
Aaron R Kaufman
PhD Candidate, Harvard University
Department of Government
818.263.5583
Hi All!
There will be no Applied Statistics presentation this week due to
Thanksgiving break. We'll resume on 12/2 with *Edo Airoldi*.
Enjoy the break, and I hope to see you all next week!
-Aaron
--
Aaron R Kaufman
PhD Candidate, Harvard University
Department of Government
818.263.5583
Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming *Luke
Miratrix*, Assistant Professor of Education at Harvard University. He will
be presenting work entitled *Estimating and assessing treatment effect
variation in large-scale randomized trials with randomization
inference**. *This
is joint work with Avi Feller and Peng Ding. Please find the abstract
below and on the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/presentations…>
.
As usual, we will meet in CGIS Knafel Room 354 from noon to 1:30pm, and
lunch will be provided. See you all there! To view previous Applied
Statistics presentations, please visit the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/videos>.
-- Aaron Kaufman
Title: Estimating and assessing treatment effect variation in large-scale
randomized trials with randomization inference
Recent literature has underscored the critical role of treatment effect
variation in estimating and understanding causal effects. This approach,
however, is in contrast to much of the foundational research on causal
inference; Neyman, for example, avoided such variation through his focus on
the average treatment effect (ATE) and his definition of the confidence
interval. We extend the Neymanian framework to explicitly allow both for
treatment effect variation explained by covariates, known as the systematic
component, and for unexplained treatment effect variation, known as the
idiosyncratic component. This perspective enables estimation and testing of
impact variation without imposing a model on the marginal distributions of
potential outcomes, with the workhorse approach of regression with
interaction terms being a special case. Our approach leads to two practical
results. First, estimates of systematic impact variation give sharp bounds
on overall treatment variation as well as bounds on the proportion of total
impact variation explained by a given model---this is essentially an R^2
for treatment effect variation. Second, by using covariates to partially
account for the correlation of potential outcomes, we sharpen the bounds on
the variance of the unadjusted average treatment effect estimate itself. As
long as the treatment effect varies across observed covariates, these
bounds are sharper than the current sharp bounds in the literature. We
demonstrate these ideas on the Head Start Impact Study, a large randomized
evaluation in educational research, showing that these results are
meaningful in practice.
Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming *Manuel
Gomez Rodriguez*, tenure-track research group director at the Max Planck
Institute for Software Systems. He will be presenting work entitled *COEVOLVE:
A Joint Point Process Model for Information Diffusion and Network
Co-evolution**.* Please find the abstract below and on the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/presentations…>
.
Manuel is pioneering an educational and research program at Max Planck in
computational social sciences, and is very excited to learn from our
experiences. Additionally, his work in machine learning, especially
learning processes over networks, has received numerous awards recently.
Those interested in meeting with Manuel are welcome to sign up here
<https://docs.google.com/spreadsheets/d/1U027d0qTmZIjw6HEf1S8rK7XTMNKt8LCldu…>
.
As usual, we will meet in CGIS Knafel Room 354 from noon to 1:30pm, and
lunch will be provided. See you all there! To view previous Applied
Statistics presentations, please visit the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/videos>.
-- Aaron Kaufman
Title: COEVOLVE: A Joint Point Process Model for Information Diffusion and
Network Co-evolution
Information diffusion in online social networks is affected by the
underlying network topology, but it also has the power to change it. Online
users are constantly creating new links when exposed to new information
sources, and in turn these links are alternating the way information
spreads. However, these two highly intertwined stochastic processes,
information diffusion and network evolution, have been predominantly
studied separately, ignoring their co-evolutionary dynamics. In this talk,
we introduce a temporal point process model, COEVOLVE, for such joint
dynamics, allowing the intensity of one process to be modulated by that of
the other. This model allows us to efficiently simulate interleaved
diffusion and network events, and generate traces obeying common diffusion
and network patterns observed in real-world networks. Furthermore, we also
develop a convex optimization framework to learn the parameters of the
model from historical diffusion and network evolution traces. We
experimented with both synthetic data and data gathered from Twitter, and
show that our model provides a good fit to the data as well as more
accurate predictions than alternatives.
--
Aaron R Kaufman
PhD Candidate, Harvard University
Department of Government
818.263.5583
Hi everyone,
This week's speaker, *Albert-Laszlo Barabasi*, as announced that his talk
will be titled *Network Science: From structure to control. *The abstract
is below. I look forward to seeing you all tomorrow!
-Aaron
Abstract: Systems as diverse as the world wide web, Internet or the cell
are described by highly interconnected networks with amazingly complex
topology. Recent studies indicate that these networks are the result of
self-organizing processes governed by simple but generic laws, resulting in
architectural features that makes them much more similar to each other than
one would have expected by chance. I will discuss the order characterizing
our interconnected world and its implications to network robustness, and
control. Indeed, while control theory offers mathematical tools to steer
engineered and natural systems towards a desired state, we lack a framework
to control complex self-organized systems. I will discuss a recently
developed analytical framework to study the controllability of an arbitrary
complex directed network, identifying the set of driver nodes whose
time-dependent control can guide the system’s dynamics.
--
Aaron R Kaufman
PhD Candidate, Harvard University
Department of Government
818.263.5583
Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming
*Albert-Laszlo
Barabasi*, Director of the Center for Complex Network Research and
Distinguished University Professor at Northeastern University. He holds
appointments in the Department of Physics and the College of Computer and
Information Science, as well as at Harvard Medical School and Brigham and
Women Hospital's Channing Division of Network Science. He has indicated
that he would like his presentation's title and abstract to be a surprise!
As usual, we will meet in CGIS Knafel Room 354 from noon to 1:30pm, and
lunch will be provided. See you all there! To view previous Applied
Statistics presentations, please visit the website
<http://projects.iq.harvard.edu/applied.stats.workshop-gov3009/videos>.
-- Aaron Kaufman
--
Aaron R Kaufman
PhD Candidate, Harvard University
Department of Government
818.263.5583
