Colin B. Fogarty

Colin B. Fogarty

I am an Assistant Professor of Operations Research and Statistics at the MIT Sloan School of Management, and I am an affiliated faculty at the Operations Research Center at MIT. Much of my work pertains to the design and analysis of observational studies while assessing the robustness of a study's findings to unobserved biases. I am also interested in experimental design, and applications of statistics to medicine and public health.

Before joining MIT, I completed my Ph.D. in Statistics at the Wharton School of the University of Pennsylvania. I was advised by Professor Dylan S. Small.

Education

Ph.D., Statistics, The Wharton School, University of Pennsylvania, 2016

A.B., Statistics, Harvard University, 2011

Publications and Submitted Papers

Heng, S., Kang. H., Small, D., and Fogarty, C. Increasing Power for Observational Studies of
Aberrant Response: An Adaptive Approach. Submitted.

Keele, L., Small, D., Hsu, J., and Fogarty, C. Patterns of Effects and Sensitivity Analysis for Differences-in-Differences. Submitted.

Cohen, P., Olson, M., and Fogarty, C. Multivariate One-Sided Testing in Matched Observational Studies as an Adversarial Game. Revision Invited.

Fogarty, C. (2019+). Studentized Sensitivity Analysis for the Sample Average Treatment Effect in Paired Observational Studies. Journal of the American Statistical Association, to appear.

Fogarty, C. and Hasegawa, R. (2019). Extended Sensitivity Analysis for Heterogeneous Unmeasured Confounding with an Application to Sibling Studies of Returns to Education. Annals of Applied Statistics, 13 (2), 767-796.

Sharifi-Malvajerdi, S., Zhu, F., Fogarty, C., Fay, M., Fairhurst, R., Flegg, J., Stepniewska, K., and Small, D. (2019). Malaria Parasite Clearance Rate Regression: An R Software Package for a Bayesian Hierarchical Regression Model. Malaria Journal, 18:4.

Fogarty, C. (2018). Regression-Assisted Inference for the Average Treatment Effect in Paired Experiments. Biometrika, 105 (4), 994–1000.

Fogarty, C. (2018). On Mitigating the Analytical Limitations of Finely Stratified Experiments. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 80, 1035-1056.

Fogarty, C., Shi, P., Mikkelsen, M., and Small, D. (2017). Randomization Inference and Sensitivity Analysis for Composite Null Hypotheses with Binary Outcomes in Matched Observational Studies. Journal of the American Statistical Association,112 (517), 321-331

Fogarty, C. and Small, D. (2016). Sensitivity Analysis for Multiple Comparisons in Matched Observational Studies
through Quadratically Constrained Linear Programming. Journal of the American Statistical Association,111 (516), 1820-1830

Fogarty, C., Mikkelsen, M., Gaieski, D., and Small, D. (2016). Discrete Optimization for Interpretable Study Populations and Randomization Inference in an Observational Study of Severe Sepsis Mortality. Journal of the American Statistical Association,111 (514), 447-458

Fogarty, C., Small, D., and Gastwirth, J. (2016). Discussion of `Perils and Potentials of Self-Selected Entry to Epidemiological Studies and Surveys' by Niels Keiding and Thomas A. Louis. Journal of the Royal Statistical Society: Series A (Statistics in Society), 179 (2), 357-358

Fogarty, C., Fay, M., Flegg, J., Stepniewska, K., Fairhurst, R. and Small, D. (2015). Bayesian Hierarchical Regression on Clearance Rates in the Presence of “Lag” and “Tail” Phases with an Application to Malaria Parasites. Biometrics, 71, 751-759.

Fogarty, C. and Small, D. (2014). Equivalence Testing for Functional Data with an Application to Comparing Pulmonary Function Devices. Annals of Applied Statistics, 8, 2002-2026.

Papers in Preparation

Fogarty, C. Near Sufficiency and Imperfect Matching in Observational Studies with Hidden Bias.

Cohen, P. and Fogarty, C. Bootstrap Prepivoting in Finite Population Casual Inference.

Fogarty, C., Keele, L., and Lee, K. Effect Heterogeneity and Omitted Variables in Paired Instrumental Variable Studies

Awards and Honors

Biometrics Early-Stage Investigator Award (2018)

Awarded by the Biometrics Section of the American Statistical Association for the paper ``Studentized Sensitivity Analysis for the Sample Average Treatment Effect in Paired Observational Studies.''

Tom Ten Have Award (2017)

Awarded at the 2017 Atlantic Causal Inference Conference for "exceptionally creative or skillful research on causal inference" for the papers ``On Mitigating the Analytical Limitations of Finely Stratified Experiments" and ``Regression Assisted Inference for the Average Treatment Effect in Paired Experiments."

Statistics in Epidemiology Young Investigator Award (2016)

Awarded by the American Statistical Association section on Statistics in Epidemiology for the paper ``Sensitivity Analysis for Multiple Comparisons in Matched Observational Studies
through Quadratically Constrained Linear Programming."

J. Parker Bursk Memorial Prize (2015)

Awarded by the Statistics Department at the Wharton School for excellence in research.

Donald S. Murray Prize (2015)

Awarded by the Statistics Department at the Wharton School for excellence in teaching.

Winkelman Fellowship (2013-2016)

Awarded to rising 3rd-year doctoral students who have shown substantial academic job potential across all departments at Wharton.

Teaching

MIT

Fall 2016, 2017: 15.S15, Readings in Statistics (Ph.D.)

Summer 2017, 2018: 15.087/15.S14, Engineering Statistics and Data Science (MBA/MS)

Spring 2017, 2018: 15.075, Statistical Thinking and Data Analysis (Undergraduate)

Wharton

Summer 2014: Stat 101, Introductory Business Statistics (Undergraduate)

Software

Note: many of the functions below require the R package for Gurobi, a solver which is freely available for academic use. Please update to version 8.0.0 of Gurobi before using these functions

bhrcr is an R package for Bayesian hierarchical regression on clearance rates as developed in Fogarty et al. (2015).

StudentizedSensitivity.R performs a sensitivity analysis for the sample average treatment effect without assuming constant effects for paired observational studies.

maxbox.R defines an interpretable study population in an observational study based on a few important covariates as the solution to the ``maximal box" problem.

compositeBinary.R conducts randomization inference and a sensitivity analysis in a matched observational study with binary outcomes for the risk difference (average treatment effect), risk ratio, and the effect ratio.

sensitivitySimple.R performs a sensitivity analysis with continuous outcomes by solving a quadratic integer program.

multiCompareFunctions.R provides functions for performing a sensitivity analysis for the overall truth of multiple hypotheses through solving a quadratically constrained linear program. This script demonstrates how to use the functions by reproducing the sensitivity analysis conducted in our paper on the effect of smoking on naphthalene levels. Here is the data on naphthalene exposure in smokers and nonsmokers.

Contact

Colin B. Fogarty

E62-582

100 Main St.

Cambridge, MA 02142

cfogarty (at) mit (dot) edu