Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming *Miguel
Hernan*, a Professor of Epidemiology at the Harvard School of Public
Health. He will be presenting work entitled *Comparative effectiveness of
dynamic treatment strategies: The renaissance of the parametric g-formula*.
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 and lunch will be provided.
See you all there!
-- Anton
Title: Comparative effectiveness of dynamic treatment strategies: The
renaissance of the parametric g-formula
Abstract: Causal questions about the comparative effectiveness and safety
of health-related interventions are becoming increasingly complex. Decision
makers are now often interested in the comparison of interventions that are
sustained over time and that may be personalized according to the
individuals’ time-evolving characteristics. These dynamic treatment
strategies cannot be adequately studied by using conventional analytic
methods that were designed to compare “treatment” vs. “no treatment”. The
parametric g-formula was developed by Robins in 1986 with the explicit goal
of comparing generalized treatment strategies sustained over time. However,
despite its theoretical superiority over conventional methods, the
parametric g-formula was rarely used for the next 25 years. Rather, the
development of causal inference methods for longitudinal data with
time-varying treatments focused on semiparametric approaches. In recent
years, interest in the parametric g-formula is growing and the number of
its applications increasing. This talk will review the parametric
g-formula, the conditions for its applicability, its practical advantages
and disadvantages compared with semiparametric methods, and several real
world implementations for comparative effectiveness research.
Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming *Fabrizia
Mealli*, Professor of Statistics, Informatics and Applications at the
University of Florence and Visiting Professor of Statistics at Harvard. She
will be presenting work entitled *Evaluating the effect of university
grants on student dropout: Evidence from a regression discontinuity design
using Bayesian principal stratification analysis*. 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 and lunch will be provided.
See you all there!
-- Anton
Title: Evaluating the effect of university grants on student dropout:
Evidence from a regression discontinuity design using Bayesian principal
stratification analysis
Abstract: Regression discontinuity (RD) designs are often interpreted as
local randomized experiments: a RD design can be considered as a randomized
experiment for units with a realized value of a so-called forcing variable
falling around a pre-fixed threshold. Motivated by the evaluation of
Italian university grants, we consider a fuzzy RD design where the receipt
of the treatment is based on both eligibility criteria and a voluntary
application status. Resting on the fact that grant application and grant
receipt statuses are post-assignment (post-eligibility) intermediate
variables, we use the principal stratification framework to define causal
estimands within the Rubin Causal Model. We propose a probabilistic
formulation of the assignment mechanism underlying RD designs, by
re-formulating the Stable Unit Treatment Value Assumption (SUTVA) and
making an explicit local overlap assumption for a subpopulation around
thethreshold. A local randomization assumption is invoked instead of more
standard continuity assumptions. We also develop a model-based Bayesian
approach to select the target subpopulation(s) with adjustment for multiple
comparisons, and to draw inference for the target causal estimands in this
framework. Applying the method to the data from two Italian universities,
we find evidence that university grants are effective in preventing
students from low-income families from dropping out of higher education.
Joint work with Fan Li and Alessandra Mattei
Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming *James
Robins*, Mitchell L. and Robin LaFoley Dong Professor of Epidemiology in
the Departments of Epidemiology and Biostatistics at Harvard University. He
will be presenting work entitled *The Foundations of Statistics and Its
Implications for Current Methods for Causal Inference from Observational
and Randomized Trial Data*. 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 and lunch will be provided.
See you all there!
-- Anton
Title: The Foundations of Statistics and Its Implications for Current
Methods for Causal Inference from Observational and Randomized Trial Data
Abstract: The foundations of statistics are the fundamental conceptual
principles that underlie statistical methodology and distinguish statistics
from the highly related fields of probability and mathematics. Examples of
foundational concepts include ancillarity, the conditionality principle,
the likelihood principle, statistical decision theory, the weak and strong
repeated sampling principle, coherence and even the meaning of probability
itself. In the 1950s and 1960s, the study of the foundations of statistics
held an important place in the field. However its central role faded with
the revolution in computing that offered the ability to actually do more
than just philosophize about how to analyze complex high dimensional data.
I discuss how these principles both inform and are informed by modern
approaches to causal analysis. Among other examples, I discuss from a
foundational perspective are (i) methods for model and/or covariate
selection including the issue of whether detailed balance on covariates is
needed after one stratifies on the true or estimated propensity, (ii) the
conflict between the minimization of MSE versus accuracy of confidence
intervals as inferential goals and (iii) the question of a whether
principled Baysesian inference must ignore the propensity score even when
it is known.
Hi everyone!
This week at the Applied Statistics Workshop we will be welcoming
Maximilian Kasy, an Assistant Professor of Economics at Harvard University.
He will be presenting work entitled *Why experimenters should not
randomize, and what they should do instead*. 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 and lunch will be provided.
See you all there!
-- Anton
Title: Why experimenters should not randomize, and what they should do
instead
Abstract: This paper discusses experimental design for the case that (i) we
are given a distribution of covariates from a pre-selected random sample,
and (ii) we are interested in the average treatment effect (ATE) of some
binary treatment. We show that in general there is a unique optimal
non-random treatment assignment if there are continuous covariates. We
argue that experimenters should choose this assignment. The optimal
assignment minimizes the risk (e.g., expected squared error) of treatment
effects estimators. We provide explicit expressions for the risk, and
discuss algorithms which minimize it. The objective of controlled trials is
to have treatment groups which are similar a priori (balanced), so we can
“compare apples with apples.” The expressions for risk derived in this
paper provide an operationalization of the notion of balance. The intuition
for our non-randomization result is similar to the reasons for not using
randomized estimators - adding noise can never decrease risk. The formal
setup we consider is decision-theoretic and nonparametric. In simulations
and an application to project STAR we find that optimal designs have mean
squared errors of up to 20% less than randomized designs and up to 14% less
than stratified designs.