RAND Statistics Seminar Series

The RAND Statistics Group holds periodic seminars that are open to all RAND employees and members of the public. The chairs of the seminar series are Brian Vegetabile and Joshua Snoke. Members of the public who would like to attend should contact Naomi Hale in advance.

If you would like to meet with any of the speakers, or request a paper, please contact Naomi Hale.

Upcoming Seminars | Past Seminars

Upcoming Seminars

Statistics and the Fair Administration of Justice

Presented by Hal Stern, University of California, Irvine

March 25, 2020
Time: 12:00 – 1:30 p.m. PT
Host location: Santa Monica
Other locations: Pittsburgh

Abstract: Statistics has emerged as a critical topic in ongoing discussions regarding the use of science to assess forensic evidence. A 2009 National Academies report on forensic science and a subsequent 2016 report by the President’s Council of Advisers on Science and Technology raised questions about the scientific underpinnings for the analysis of a number of types of forensic evidence. Misapplication of forensic science has been identified as a contributing factor in nearly half of 362 cases in which DNA helped exonerate wrongly-convicted individuals. For these reasons there has been an increased focus on evaluating the ways in which evidence is analyzed, interpreted and reported with an eye towards providing more scientifically justified methods. There are three common approaches to the analysis and interpretation of forensic evidence: (1) forensic conclusions as expert opinion; (2) two-stage procedures (determination of similarity of known/questioned items followed by an assessment of significance); (3) likelihood ratios. The first of these, forensic conclusions as expert opinion, has been the standard approach in the pattern comparison disciplines (e.g., fingerprints, firearms, handwriting) but is now being questioned. The logical and statistical issues associated with each of the approaches are discussed in the context of current research.

Nonparametric causal effects based on incremental propensity score interventions

Presented by Edward Kennedy, Carnegie Mellon University

April 30, 2020
Time: 10:30 a.m. – 12:00 p.m. PT
Host location: Pittsburgh

Most work in causal inference considers deterministic interventions that set each unit’s treatment to some fixed value. However, when some units have zero chance of being treated at that level (i.e., positivity violations) these interventions can lead to non-identification, inefficiency, and effects with little practical relevance. Further, corresponding effects in longitudinal studies are highly sensitive to the curse of dimensionality, resulting in widespread use of unrealistic parametric models. We propose a novel solution to these problems: incremental interventions that shift propensity score values rather than set treatments to fixed values. Incremental interventions have several crucial advantages. First, they avoid positivity assumptions entirely. Second, they require no parametric assumptions and yet still admit a simple characterization of longitudinal effects, independent of the number of timepoints. For example, they allow longitudinal effects to be visualized with a single curve instead of lists of coefficients. After characterizing incremental interventions and giving identifying conditions for corresponding effects, we also develop general efficiency theory, propose efficient nonparametric estimators that can attain fast convergence rates even when incorporating flexible machine learning, and propose a bootstrap-based confidence band and simultaneous test of no treatment effect. Finally, we explore finite-sample performance via simulation, and apply the methods to study time-varying sociological effects of incarceration on entry into marriage.

Past Seminars

Prediction, Estimation, and Attribution

Presented by Bradley Efron, Stanford University

Thursday, November 6, 2019
Time: 12:00 p.m. – 1:30 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Alternatives to p-values

Presented by Daniel Jeske, University of California Riverside

Thursday, September 25, 2019
Time: 12:00 p.m. – 1:30 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Deep Learning: A Synthesis from Probabilistic Foundations

Presented by Eric Nalisnick, University of Cambridge and Deep Mind

Thursday, June 18, 2019
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Making Sense of Sensitivity: Extending Omitted Variable Bias

Presented by Carlos Cinelli, UCLA
Co-Sponsored with Center for Causal Inference (CCI)

Thursday, May 23, 2019
Time: 12:00 p.m. – 1:30 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Causal Inference in the Presence of Interference

Presented by Michael Hudgens, University of North Carolina
Co-Sponsored with Center for Causal Inference (CCI)

Thursday, November 15, 2018
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Statistical Methods Research done as Science rather than Mathematics

Presented by Jim Hodges, University of Minnesota, Division of Biostatistics, School of Public Health

Thursday, May 18, 2017
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Assessing Time-Varying Causal Effect Moderation in Mobile Health

Presented by Daniel Almirall, University of Michigan
Co-Sponsored with Center for Causal Inference (CCI)

Thursday, January 26, 2017
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Multivariate Multilevel Models for Health Care Quality Data

Presented by Alan Zaslavsky, Harvard Medical School, Professor of Health Care Policy (statistics)

Thursday, November 11, 2016
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, DC

Personalized Biostatistics, Small Data, and N-of-1 Trials

Presented by Naihua Duan, Columbia University, Division of Biostatistics, Department of Psychiatry

Thursday, October 10, 2016
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, DC

A synthetic estimator to improve estimation of causal effects in clinical trials with noncompliance

Presented by Joseph (Joey) Antonelli, Harvard Biostatistics, Postdoctoral Fellow
Co-Sponsored by Center for Causal Inference (CCI)

Thursday, August 25, 2016
Time: 12:00 p.m. – 1:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

Endogenous Stratification in Randomized Experiment

Presented by Alberto Abadie, MIT

Thursday, July 18, 2016
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, D.C.

An Experiment of Experimentalism: Does Peer Review Work?

Presented by Daniel Ho, Stanford University

Thursday, June 30, 2016
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, DC

Principal stratification in the Twilight Zone: Weakly separated components in finite mixture models

Presented by Avi Feller, University of California, Berkeley

Thursday, May 19, 2016
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, DC

Collaborative Double Robustness and The Connection to Data-Adaptive Nuisance Model Selection in Causal Inference

Presented by Mireille Schnitzer, Assistant Professor of Biostatistics, University of Montreal

Thursday, March 10, 2016
Time: 10:30 a.m. – 12:00 p.m. Pacific
Host Location: Santa Monica
Other Locations: Pittsburgh, Washington, DC

RAND Statistics Seminar Series

Upcoming Seminars | Past Seminars

Upcoming Seminars

Statistics and the Fair Administration of Justice

Nonparametric causal effects based on incremental propensity score interventions

Past Seminars

Prediction, Estimation, and Attribution

Alternatives to p-values

Deep Learning: A Synthesis from Probabilistic Foundations

Making Sense of Sensitivity: Extending Omitted Variable Bias

Causal Inference in the Presence of Interference

Statistical Methods Research done as Science rather than Mathematics

Assessing Time-Varying Causal Effect Moderation in Mobile Health

Multivariate Multilevel Models for Health Care Quality Data

Personalized Biostatistics, Small Data, and N-of-1 Trials

A synthetic estimator to improve estimation of causal effects in clinical trials with noncompliance

Endogenous Stratification in Randomized Experiment

An Experiment of Experimentalism: Does Peer Review Work?

Principal stratification in the Twilight Zone: Weakly separated components in finite mixture models

Collaborative Double Robustness and The Connection to Data-Adaptive Nuisance Model Selection in Causal Inference

Estimating Causal Effects of Continuous Exposures

Explained Variation in Correlated Survival Data

Proof in Medicine: The Contributions of Jerome Cornfield

Penalized and Constrained Regression

Design and Estimation for Respondent Driven Sampling

Essential Ideas of Causal Inference in Experiments and in Observational Studies

Robust Mixed-effects Model for Clustered Failure Time Data: Application to Huntington's Disease Event Measures

Propensity Score Methods in the Context of Covariate Measurement Error

A Gaussian Process Model for Estimating Within-Subject Volatility in Indices of Protein-Energy Malnutrition Among End-Stage Renal Disease Patients

Adaptive Piecewise Polynomial Estimation via Trend Filtering

Non-Local Priors for High-Dimensional Estimation

Cardiovascular Event Risk Dynamics Over Time in Older Patients on Dialysis: A Generalized Multiple-Index Varying Coefficient Model Approach

Missing Data as a Causal Inference problem

Estimating the Comparative Effectiveness of Screening Tests

Exits on the Road to Serfdom

Data-adaptive Approaches to Causal Effect Estimation

Systematic Approaches to Personalized Treatment Selection

Modeling Networks When Data Is Missing or Sampled

Markov Chain Monte Carlo: Can We Trust the Third Significant Figure?

Effect of multivariate outlier detection & imputation on poverty & inequality indicators

Sparse Nonparametric Bayesian Learning from Big Data

Examining moderated effects of additional adolescent substance use treatment: Structural nested mean model estimation using inverse-weighted regression-with-residuals

Tweedie's Formula and Selection Bias

Effective Semiparametric Modeling for Ultra-Sparse, Unsynchronized and Imprecise Longitudinal Data

Sparse Functional Data with an Irregular Periodic Component: An Application to Psychiatric Data

Multiplicative Algorithms: Why Do They Converge and How to Make Them Faster

Modern Experimental Design Methods and How to Use Them

What's New in Causal Inference: From Propensity Scores and Mediation to External Validity and Surrogate Endpoints

Context Modeling for Detecting Cis-Regulatory Signals

Valued Ties Tell Fewer Lies: Why Not To Dichotomize Network Edges With Thresholds

Identification of Social Interactions

The Importance of Scale for Spatial-Confounding Bias and Precision of Spatial Regression Estimators

A Coverage Approach to Evaluating Mean Square Error

Measuring HIV Incidence: Approaches and Challenges

High Dimensional Classification Using Feature Annealed Independence Rules

Propensity Score Matching to Recover Latent Experiments

Exploring the Housing Crisis with ggplot2 and plyr in R

Lessons from the Netflix Prize

Using the Whole Cohort in Two-Phase Study Designs

Data Large-Scale Prediction Problems

Everything is Dangerous: A Controversy

Information, Knowledge, and Truth: Replacing Secrecy, Ignorance, and Myth in the Wake of Atrocity

Proxy Pattern-Mixture Analysis for Survey Nonresponse

Accounting for Complex Sample Designs via Mixture Models

Visualizing Hierarchical Cluster Structure via Density-Based Linkage Methods and Component Trees

Data Assimilation for Epidemiologic Air Exposure Assessment

Statistics and Public Policy

Analysis of Gene Expression Leukemia Data using Split-Merge Markov Chain Monte Carlo

Causal Inference for Randomized Trials of Two Active Treatments Subject to Non-compliance

Spectral Analysis of Faint Astronomical Objects: Bayesian Modeling, Computation, and Inference

A Stimulus-Locked VAR Model for fMRI Event-Related Connectivity Analyses

Robust Standard Errors and Experimental Design

Simultaneous Inference: When Should Hypothesis Testing Problems Be Combined?

Secure Logistic Regression with Distributed Databases

Enabling Knowledge Networks in 21st Century Organizational Forms: From Disasters to WoW

Improving Efficiency of Inferences in Randomized Clinical Trials Using Auxiliary Covariates

A Multivariate Nongaussian Bayesian Spatial Modeling Framework for Hurricane Surface Wind Fields

"Proof" in Medicine: The Role of Research Synthesis

Visualization Databases for Loseless Analysis of Complex Data Sets

Clustering Functional Data: Methods and Applications

Identification of the Variance Components in the
General Two-Variance Linear Model

Responsive Design for Household Surveys:
Tools for Actively Controlling Survey Nonresponse and Costs