Siyu Heng
Siyu Heng
Scroll
Assistant Professor of Biostatistics
-
Professional overview
-
Siyu Heng, PhD is an Assistant Professor in the Department of Biostatistics, with an interest in both methodology research and applied research. His areas of expertise are in causal inference, health data science, observational studies, randomized trials, sensitivity analysis, instrumental variables, measurement error, and in survey data and their applications in public health.
Dr. Heng’s research has been published in the Journal of the Royal Statistical Society and in Physical Review, among others. He has been recognized with several awards, including the IPUMS Global Health Research Award for the Best Student Paper; the Lawrence D. Brown Best Paper Award; the ASA Mental Health Statistics Section Student Paper Award; the ENAR Distinguished Student Paper Award; the NESS Student Research Award, and the Wellcome Trust Data Reuse Prize.
Dr. Heng received his PhD in applied mathematics and computational science from the University of Pennsylvania, and his BA in statistics from Nanjing University.
-
Education
-
PhD Candidate, Applied Mathematics and Computational Science (Statistics Track) University of PennsylvaniaBS, Mathematics, Nanjing University
-
Honors and awards
-
IPUMS Global Health Research Award for the Best Student Paper, Integrated Public Use Microdata Series (2021)IMS Hannan Graduate Student Travel Award, Institute of Mathematical Statistics (2021)ASA Mental Health Statistics Section Student Paper Award, American Statistical Association Section on Mental Health Statistics (2021)ENAR Distinguished Student Paper Award, International Biometric Society Eastern North American Region (2021)Wellcome Trust Data Reuse Prize: Malaria, Wellcome Trust (2019)Benjamin Franklin Fellowship, University of Pennsylvania School of Arts and Sciences (201620172018)
-
Areas of research and study
-
Causal InferenceEpidemiologyGlobal HealthHealth EquityInstrumental VariablesObservational StudiesPublic Health PolicyRandomized ExperimentationSocial Sciences
-
Publications
Publications
DELTA: Dual Consistency Delving with Topological Uncertainty for Active Graph Domain Adaptation
AbstractWang, P., Cao, Y., Russell, C., Shen, Y., Luo, J., Zhang, M., Heng, S., & Luo, X. (n.d.).Publication year
2025Journal title
Transactions on Machine Learning ResearchVolume
2025Page(s)
1-26AbstractGraph domain adaptation has recently enabled knowledge transfer across different graphs. However, without the semantic information on target graphs, the performance on target graphs is still far from satisfactory. To address the issue, we study the problem of active graph domain adaptation, which selects a small quantitative of informative nodes on the target graph for extra annotation. This problem is highly challenging due to the complicated topological relationships and the distribution discrepancy across graphs. In this paper, we propose a novel approach named Dual Consistency Delving with Topological Uncertainty (DELTA) for active graph domain adaptation. Our DELTA consists of an edge-oriented graph subnetwork and a path-oriented graph subnetwork, which can explore topological semantics from complementary perspectives. In particular, our edge-oriented graph subnetwork utilizes the message passing mechanism to learn neighborhood information, while our path-oriented graph subnetwork explores high-order relationships from sub-structures. To jointly learn from two subnetworks, we roughly select informative candidate nodes with the consideration of consistency across two subnetworks. Then, we aggregate local semantics from its K-hop subgraph based on node degrees for topological uncertainty estimation. To overcome potential distribution shifts, we compare target nodes and their corresponding source nodes for discrepancy scores as an additional component for fine selection. Extensive experiments on benchmark datasets demonstrate that DELTA outperforms various state-of-the-art approaches. The code implementation of DELTA is available at https://github.com/goose315/DELTA.Effects of behavioral intervention components to increase COVID-19 testing for African American/Black and Latine frontline essential workers not up-to-date on COVID-19 vaccination: Results of an optimization randomized controlled trial
AbstractGwadz, M., Heng, S., Cleland, C. M., Strayhorn, J., Robinson, J. A., Serrano, F. G. B., Wang, P., Parameswaran, L., & Chero, R. (n.d.).Publication year
2025Journal title
Journal of Behavioral MedicineAbstractRacial/ethnic disparities in COVID-19, including incidence, hospitalization, and death rates, are serious and persistent. Among those at highest risk for COVID-19 and its adverse effects are African American/Black and Latine (AABL) frontline essential workers in public-facing occupations (e.g., food services, retail). Testing for COVID-19 in various scenarios (when exposed or symptomatic, regular screening testing) is an essential component of the COVID-19 control strategy in the United States. However, AABL frontline workers have serious barriers to COVID-19 testing at the individual (insufficient knowledge, distrust, cognitive biases), social (norms), and structural levels of influence (access). Thus, testing rates are insufficient and interventions are needed. The present study is grounded in the multiphase optimization strategy (MOST) framework. It tests the main and interaction effects of a set of candidate behavioral intervention components to increase COVID-19 testing rates in this population. The study enrolled adult AABL frontline essential workers who were not up-to-date on COVID-19 vaccination nor recently tested for COVID-19. It used a factorial design to examine the effects of candidate behavioral intervention components, where each component was designed to address a specific barrier to COVID-19 testing. All participants received a core intervention comprised of health education. The candidate components were motivational interviewing counseling (MIC), a behavioral economics intervention (BEI), peer education (PE), and access to testing (either self-test kits [SK] or a navigation meeting [NM]). The primary outcome was COVID-19 testing in the follow-up period. Participants were assessed at baseline, randomly assigned to one of 16 experimental conditions, and assessed six- and 12-weeks later. The study was carried out in English and Spanish. We used a logistic regression model and multiple imputation to examine the main and interaction effects of the four factors (representing components): MIC, BEI, PE, and Access. We also conducted a sensitivity analysis using the complete case analysis. Participants (N = 438) were 35 years old on average (SD = 10). Half identified as men/male (52%), and 48% as women/female/other. Almost half (49%) were African American/Black, and 51% were Latine/Hispanic (12% participated in Spanish). A total of 32% worked in food services. Attendance in components was very high (~ 99%). BEI had positive effect on the outcome (OR = 1.543; 95% CI: [0.977, 2.438]; p-value = 0.063) as did Access, in favor of SK (OR = 1.351; 95% CI: [0.859, 2.125]; p-value = 0.193). We found a three-way interaction among MIC*PE*Access (OR: 0.576; 95% CI: [0.367, 0.903]; p-value = 0.016): when MIC was present, SK tended to increase COVID testing when PE was not present. The study advances intervention science and takes the first step toward creating an efficient and effective multi-component intervention to increase COVID-19 testing rates in AABL frontline workers.Parent-daughter agreement about HPV vaccination status in Kenya and Malawi
AbstractMoucheraud, C., Ochieng, E., Kweka, A., Wang, P., Xie, S., Ototo, J., Golub, G., Kapindo, E., Banda, E., Abdillahi, H., Szilagyi, P. G., & Heng, S. (n.d.).Publication year
2025Journal title
VaccineVolume
55AbstractBackground: As more countries introduce the HPV vaccine, it is important to understand the validity of vaccination measures. This is especially true in low- and middle-income countries (LMICs) where public health monitoring of vaccination data may have delays or gaps, so alternative measurement approaches are often necessary. Parental report is a common approach for measuring routine childhood vaccination, but it has not been evaluated for HPV vaccination in LMICs. Methods: We conducted household surveys in Kenya (n = 146) and Malawi (n = 98) with parents/guardians and their daughters who were age-eligible for HPV vaccination. We compared parents'/guardians' reports of HPV vaccination status to daughters' reports; the latter was assumed to be the “gold standard” measure. Results: 88 % of Kenyan parents/guardians and 82 % of Malawian parents/guardians agreed with their daughters' reported HPV vaccination status. It was more common for parents/guardians to under-report (i.e., to say their daughter was unvaccinated but the girl said she had received dose(s)) than the inverse. Agreement with one's daughter was higher among parents/guardians who reported data from vaccination cards versus using recall, and among parents/guardians who expressed more versus less confidence in their knowledge. We did not find many differences in accuracy of report by parent/guardian characteristics, although in Kenya there were small and statistically significant negative associations with parental age, household income, and more girls in the household (the latter was also significantly negatively associated with report accuracy in Malawi). Conclusions: In countries where surveys will commonly be used to measure HPV vaccination status, we found very high agreement of parents/guardians with their daughters' reported receipt of the vaccine. These results are similar to findings from the literature about routine childhood vaccination measurement. This suggests that researchers, clinicians, and practitioners can use parent/guardian-reported HPV vaccination of their daughter as a relatively good proxy of her own reported immunization status especially in settings without universal use of vaccination cards or registries.Maximizing the reach of universal child sexual abuse prevention: Protocol for an equivalence trial
AbstractGuastaferro, K., Melchior, M. S., Heng, S., Trudeau, J., & Holloway, J. L. (n.d.).Publication year
2024Journal title
Contemporary Clinical Trials CommunicationsVolume
41AbstractBackground: Child sexual abuse (CSA) affects 1 in 5 girls and 1 in 12 boys before age 18. Universal school-based prevention programs are an effective and cost-efficient method of teaching students an array of personal safety skills. However, the programmatic reach of universal school-based programs is limited by the inherent reliance on the school infrastructure and a dearth of available alternative delivery modalities. Methods: The design for this study will use a rigorous cluster randomized design (N = 180 classrooms) to determine the equivalence of two delivery modalities of Safe Touches: as usual vs. modified. The as usual workshop will be delivered by two facilitators with live puppet skits (n = 90). Whereas, the modified workshop will be delivered by one facilitator using prerecorded skit videos (n = 90). We will determine the equivalence by measuring concept learning acquisition preworkshop to immediate postworkshop (Aim 1) and retention at 3-months postworkshop (Aim 2) among students in classrooms that receive the as usual or modified workshops. To conclude equivalence, it is imperative to also examine factors that may impact future dissemination and implementation, specifically program adoption among school personnel and implementation fidelity between the two modalities (Aim 3). Conclusion: Study findings will inform the ongoing development of effective CSA prevention programs and policy decisions regarding the sustainable integration of such programs within schools. Clinical trial registration: NCT06195852.Sensitivity analysis for matched observational studies with continuous exposures and binary outcomes
AbstractZhang, J., Small, D. S., & Heng, S. (n.d.).Publication year
2024Journal title
BiometrikaVolume
111Issue
4Page(s)
1349-1368AbstractMatching is one of the most widely used study designs for adjusting for measured confounders in observational studies. However, unmeasured confounding may exist and cannot be removed by matching. Therefore, a sensitivity analysis is typically needed to assess a causal conclusion's sensitivity to unmeasured confounding. Sensitivity analysis frameworks for binary exposures have been well established for various matching designs and are commonly used in various studies. However, unlike the binary exposure case, there still lacks valid and general sensitivity analysis methods for continuous exposures, except in some special cases such as pair matching. To fill this gap in the binary outcome case, we develop a sensitivity analysis framework for general matching designs with continuous exposures and binary outcomes. First, we use probabilistic lattice theory to show that our sensitivity analysis approach is finite population exact under Fisher's sharp null. Second, we prove a novel design sensitivity formula as a powerful tool for asymptotically evaluating the performance of our sensitivity analysis approach. Third, to allow effect heterogeneity with binary outcomes, we introduce a framework for conducting asymptotically exact inference and sensitivity analysis on generalized attributable effects with binary outcomes via mixed-integer programming. Fourth, for the continuous outcome case, we show that conducting an asymptotically exact sensitivity analysis in matched observational studies when both the exposures and outcomes are continuous is generally NP-hard, except in some special cases such as pair matching. As a real data application, we apply our new methods to study the effect of early-life lead exposure on juvenile delinquency. An implementation of the methods in this work is available in the R package doseSens.Instrumental variables: to strengthen or not to strengthen?
AbstractHeng, S., Zhang, B., Han, X., Lorch, S. A., & Small, D. S. (n.d.).Publication year
2023Journal title
Journal of the Royal Statistical Society. Series A: Statistics in SocietyVolume
186Issue
4Page(s)
852-873AbstractInstrumental variables (IVs) are extensively used to handle unmeasured confounding. However, weak IVs may cause problems. Many matched studies have considered strengthening an IV through discarding some of the sample. It is widely accepted that strengthening an IV tends to increase the power of non-parametric tests and sensitivity analyses. We re-evaluate this conventional wisdom and offer new insights. First, we evaluate the trade-off between IV strength and sample size assuming a valid IV and exhibit conditions under which strengthening an IV increases power. Second, we derive a criterion for checking the validity of a sensitivity analysis model with a continuous dose and show that the widely used Γ sensitivity analysis model, which was used to argue that strengthening an IV increases the power of sensitivity analyses in large samples, does not work for continuous IVs. Third, we quantify the bias of the Wald estimator with a possibly invalid IV and leverage it to develop a valid sensitivity analysis framework and show that strengthening an IV may or may not increase the power of sensitivity analyses. We use our framework to study the effect on premature babies of being delivered in a high technology/high volume neonatal intensive care unit.SOCIAL DISTANCING AND COVID-19: RANDOMIZATION INFERENCE FOR A STRUCTURED DOSE-RESPONSE RELATIONSHIP
AbstractZhang, B., Heng, S., Ye, T., & Small, D. S. (n.d.).Publication year
2023Journal title
Annals of Applied StatisticsVolume
17Issue
1Page(s)
23-46AbstractSocial distancing is widely acknowledged as an effective public health policy combating the novel coronavirus. But extreme forms of social distanc-ing, like isolation and quarantine, have costs, and it is not clear how much social distancing is needed to achieve public health effects. In this article we develop a design-based framework to test the causal null hypothesis and make inference about the dose-response relationship between reduction in social mobility and COVID-19 related public health outcomes. We first dis-cuss how to embed observational data with a time-independent, continuous treatment dose into an approximate randomized experiment and develop a randomization-based procedure that tests if a structured dose-response relationship fits the data. We then generalize the design and testing procedure to a longitudinal setting and apply them to investigate the effect of social distancing during the first phased reopening in the United States on public health outcomes using data compiled from Unacast™, the United States Census Bu-reau, and the County Health Rankings and Roadmaps Program. We rejected a primary analysis null hypothesis that stated the social distancing from April 27, 2020 to June 28, 2020, had no effect on the COVID-19-related death toll from June 29, 2020 to August 2, 2020 (p-value < 0.001), and found that it took more reduction in mobility to prevent exponential growth in case num-bers for nonrural counties compared to rural counties.Testing Biased Randomization Assumptions and Quantifying Imperfect Matching and Residual Confounding in Matched Observational Studies
AbstractChen, K., Heng, S., Long, Q., & Zhang, B. (n.d.).Publication year
2023Journal title
Journal of Computational and Graphical StatisticsVolume
32Issue
2Page(s)
528-538AbstractOne central goal of design of observational studies is to embed nonexperimental data into an approximate randomized controlled trial using statistical matching. Despite empirical researchers’ best intention and effort to create high-quality matched samples, residual imbalance due to observed covariates not being well matched often persists. Although statistical tests have been developed to test the randomization assumption and its implications, few provide a means to quantify the level of residual confounding due to observed covariates not being well matched in matched samples. In this article, we develop two generic classes of exact statistical tests for a biased randomization assumption. One important by-product of our testing framework is a quantity called residual sensitivity value (RSV), which provides a means to quantify the level of residual confounding due to imperfect matching of observed covariates in a matched sample. We advocate taking into account RSV in the downstream primary analysis. The proposed methodology is illustrated by re-examining a famous observational study concerning the effect of right heart catheterization (RHC) in the initial care of critically ill patients. Code implementing the method can be found in the supplementary materials.The Central Role of the Propensity Score in Sensitivity Analysis for Matched Observational Studies
AbstractHeng, S. (n.d.).Publication year
2023Journal title
Observational StudiesVolume
9Issue
1Page(s)
35-41AbstractThe propensity score, which was originally introduced in Rosenbaum and Rubin (1983), has been widely considered one of the most important concepts in the causal inference literature. This article briefly reviews some propensity score models involving both observed and unobserved covariates and discusses their applications in sensitivity analysis for matched observational studies.Bridging preference-based instrumental variable studies and cluster-randomized encouragement experiments: Study design, noncompliance, and average cluster effect ratio
AbstractZhang, B., Heng, S., MacKay, E. J., & Ye, T. (n.d.).Publication year
2022Journal title
BiometricsVolume
78Issue
4Page(s)
1639-1650AbstractInstrumental variable (IV) methods are widely used in medical research to draw causal conclusions when the treatment and outcome are confounded by unmeasured confounding variables. One important feature of such studies is that the IV is often applied at the cluster level, for example, hospitals' or physicians' preference for a certain treatment where each hospital or physician naturally defines a cluster. This paper proposes to embed such observational IV data into a cluster-randomized encouragement experiment using nonbipartite matching. Potential outcomes and causal assumptions underpinning the design are formalized and examined. Testing procedures for two commonly used estimands, Fisher's sharp null hypothesis and the pooled effect ratio (PER), are extended to the current setting. We then introduce a novel cluster-heterogeneous proportional treatment effect model and the relevant estimand: the average cluster effect ratio. This new estimand is advantageous over the structural parameter in a constant proportional treatment effect model in that it allows treatment heterogeneity, and is advantageous over the PER estimand in that it does not suffer from Simpson's paradox. We develop an asymptotically valid randomization-based testing procedure for this new estimand based on solving a mixed-integer quadratically constrained optimization problem. The proposed design and inferential methods are applied to a study of the effect of using transesophageal echocardiography during coronary artery bypass graft surgery on patients' 30-day mortality rate. R package ivdesign implements the proposed method.Hemoglobin Levels among Male Agricultural Workers: Analyses from the Demographic and Health Surveys to Investigate a Marker for Chronic Kidney Disease of Uncertain Etiology
AbstractLin, Y., Heng, S., Anand, S., Deshpande, S. K., & Small, D. S. (n.d.).Publication year
2022Journal title
Journal of Occupational and Environmental MedicineVolume
64Issue
12Page(s)
E805-E810AbstractObjective Estimate agricultural work's effect on hemoglobin (Hgb) level in men. A negative effect may indicate presence of chronic kidney disease of uncertain etiology. Methods We use Demographic and Health Surveys data from seven African and Asian countries and use matching to control for seven confounders. Results On average, Hgb levels were 0.09 g/dL lower among agricultural workers compared with matched controls. Significant effects were observed in Ethiopia, India, Lesotho, and Senegal, with effects from 0.07 to 0.30 g/dL lower Hgb level among agricultural workers. The findings were robust to multiple control groups and a modest amount of unmeasured confounding. Conclusions Men engaged in agricultural work in four of the seven countries studied have modestly lower Hgb levels. Our data support integrating kidney function assessments within Demographic and Health Surveys and other population-based surveys.Association between Transesophageal Echocardiography and Clinical Outcomes after Coronary Artery Bypass Graft Surgery
AbstractMacKay, E. J., Zhang, B., Heng, S., Ye, T., Neuman, M. D., Augoustides, J. G., Feinman, J. W., Desai, N. D., & Groeneveld, P. W. (n.d.).Publication year
2021Journal title
Journal of the American Society of EchocardiographyVolume
34Issue
6Page(s)
571-581AbstractBackground: Coronary artery bypass graft (CABG) surgery is the most widely performed cardiac surgery in the United States. Transesophageal echocardiography (TEE) is frequently used in a variety of cardiac surgical procedures, but its clinical benefit in isolated CABG surgery is unclear, and guidelines remain indeterminate. The aim of this study was to compare clinical outcomes among patients undergoing isolated CABG surgery with versus without TEE in order to test the hypothesis that TEE would be associated with improved clinical outcomes after CABG surgery. Methods: A matched retrospective cohort study was conducted among Medicare beneficiaries undergoing isolated CABG surgery with versus without intraoperative monitoring using TEE in the United States. The primary analysis was a near/far instrumental variable match that paired hospitals with similar characteristics and patient populations but with opposing probabilities for using TEE in CABG surgery. Outcomes included 30-day mortality, a composite outcome of stroke or 30-day mortality, length of hospitalization, and incidence of esophageal perforation. Results: Of 114,871 patients undergoing isolated CABG surgery, 65,471 (57%) underwent TEE and 49,400 (43%) did not. Hospital-level instrumental variable matched analysis demonstrated that among the subset of 968 matched hospitals, TEE receipt was associated with lower 30-day mortality (3.7% vs 4.9%, P <.001), a lower incidence of the composite outcome of stroke or 30-day mortality (4.5% vs 5.6%, P <.001), no difference in length of hospitalization (10.32 vs 10.52 days, P =.26), and no difference in the incidence of esophageal perforation (0.01% vs 0.01%, P =.63). These results were replicated in surgeon-level and patient-level matched-pair instrumental variable analyses, and all analyses were robust to sensitivity analyses that tested for biases introduced by unmeasured confounding. Conclusions: The findings from this study suggest that TEE may offer a clinical benefit to cardiac surgical patients undergoing isolated CABG surgery.Increasing power for observational studies of aberrant response: An adaptive approach
AbstractHeng, S., Kang, H., Small, D. S., & Fogarty, C. B. (n.d.).Publication year
2021Journal title
Journal of the Royal Statistical Society. Series B: Statistical MethodologyVolume
83Issue
3Page(s)
482-504AbstractIn many observational studies, the interest is in the effect of treatment on bad, aberrant outcomes rather than the average outcome. For such settings, the traditional approach is to define a dichotomous outcome indicating aberration from a continuous score and use the Mantel–Haenszel test with matched data. For example, studies of determinants of poor child growth use the World Health Organization’s definition of child stunting being height-for-age z-score ≤ − 2. The traditional approach may lose power because it discards potentially useful information about the severity of aberration. We develop an adaptive approach that makes use of this information and asymptotically dominates the traditional approach. We develop our approach in two parts. First, we develop an aberrant rank approach in matched observational studies and prove a novel design sensitivity formula enabling its asymptotic comparison with the Mantel–Haenszel test under various settings. Second, we develop a new, general adaptive approach, the two-stage programming method, and use it to adaptively combine the aberrant rank test and the Mantel–Haenszel test. We apply our approach to a study of the effect of teenage pregnancy on stunting.Relationship between changing malaria burden and low birth weight in sub-saharan africa: A difference-in-differences study via a pair-of-pairs approach
AbstractHeng, S., O’meara, W. P., Simmons, R. A., & Small, D. S. (n.d.).Publication year
2021Journal title
eLifeVolume
10AbstractBackground: According to the World Health Organization (WHO), in 2018, an estimated 228 million malaria cases occurred worldwide with most cases occurring in sub-Saharan Africa. Scale-up of vector control tools coupled with increased access to diagnosis and effective treatment has resulted in a large decline in malaria prevalence in some areas, but other areas have seen little change. Although interventional studies demonstrate that preventing malaria during pregnancy can reduce the rate of low birth weight (i.e. child’s birth weight <2500 g), it remains unknown whether natural changes in parasite transmission and malaria burden can improve birth outcomes. Methods: We conducted an observational study of the effect of changing malaria burden on low birth weight using data from 18,112 births in 19 countries in sub-Saharan African countries during the years 2000–2015. Specifically, we conducted a difference-in-differences study via a pair-of-pairs matching approach using the fact that some sub-Saharan areas experienced sharp drops in malaria prevalence and some experienced little change. Results: A malaria prevalence decline from a high rate (Plasmodium falciparum parasite rate in children aged 2-up-to-10 (i.e. PfPR2-10) > 0.4) to a low rate (PfPR2-10 < 0.2) is estimated to reduce the rate of low birth weight by 1.48 percentage points (95% confidence interval: 3.70 percentage points reduction, 0.74 percentage points increase), which is a 17% reduction in the low birth weight rate compared to the average (8.6%) in our study population with observed birth weight records (1.48/8.6 » 17%). When focusing on first pregnancies, a decline in malaria prevalence from high to low is estimated to have a greater impact on the low birth weight rate than for all births: 3.73 percentage points (95% confidence interval: 9.11 percentage points reduction, 1.64 percentage points increase). Conclusions: Although the confidence intervals cannot rule out the possibility of no effect at the 95% confidence level, the concurrence between our primary analysis, secondary analyses, and sensitivity analyses, and the magnitude of the effect size, contribute to the weight of the evidence suggesting that declining malaria burden can potentially substantially reduce the low birth weight rate at the community level in sub-Saharan Africa, particularly among firstborns. The novel statistical methodology developed in this article–a pair-of-pairs approach to a difference-in-differences study–could be useful for many settings in which different units are observed at different times.SHARPENING THE ROSENBAUM SENSITIVITY BOUNDS TO ADDRESS CONCERNS ABOUT INTERACTIONS BETWEEN OBSERVED AND UNOBSERVED COVARIATES
AbstractHeng, S., & Small, D. S. (n.d.).Publication year
2021Journal title
Statistica SinicaVolume
31Page(s)
2331-2353AbstractIn observational studies, it is typically unrealistic to assume that treatments are assigned randomly, even conditional on adjusting for all observed covariates. Therefore, a sensitivity analysis is often needed to examine how hidden biases due to unobserved covariates affect inferences on treatment effects. In matched observational studies, where each treated unit is matched to one or multiple untreated controls for observed covariates, the Rosenbaum bounds sensitivity analysis is one of the most popular sensitivity analysis models. We show that in the presence of interactions between observed and unobserved covariates, directly applying the Rosenbaum bounds almost inevitably exaggerates the report of sensitivity of causal conclusions to hidden bias. We give sharper odds ratio bounds to fix this deficiency. We illustrate our new method by studying the effect of a anger/hostility tendency on the risk of experiencing heart problems.Finding the strength in a weak instrument in a study of cognitive outcomes produced by Catholic high schools
AbstractHeng, S., Small, D. S., & Rosenbaum, P. R. (n.d.).Publication year
2020Journal title
Journal of the Royal Statistical Society. Series A: Statistics in SocietyVolume
183Issue
3Page(s)
935-958AbstractWe show that the strength of an instrument is incompletely characterized by the proportion of compliers, and we propose and evaluate new methods that extract more information from certain settings with comparatively few compliers. Specifically, we demonstrate that, for a fixed small proportion of compliers, the presence of an equal number of always-takers and never-takers weakens an instrument, whereas the absence of always-takers or, equivalently, the absence of never-takers strengthens an instrument. In this statement, the strength of an instrument refers to its ability to recognize and reject a false hypothesis about a structural parameter. Equivalently, the strength of an instrument refers to its ability to exclude from a confidence interval a false value of a structural parameter. This ability is measured by the Bahadur efficiency of a test that assumes that the instrument is flawless, or the Bahadur efficiency of a sensitivity analysis that assumes that the instrument may be somewhat biased. When there are few compliers, the outcomes for most people are unaffected by fluctuations in the instrument, so most of the information about the treatment effect is contained in the tail of the distribution of the outcomes. Exploiting this fact, we propose new methods that emphasize the affected portion of the distribution of outcomes, thereby extracting more information from studies with few compliers. Studies of the effects of Catholic high schools on academic test performance have used ‘being Catholic’ as an instrument for ‘attending a Catholic high school’, and the application concerns such a comparison using the US National Educational Longitudinal Study. Most Catholics did not attend Catholic school, so there are few compliers, but it was rare for non-Catholics to attend Catholic school, so there are very few always-takers.Can phoretic particles swim in two dimensions?
AbstractSondak, D., Hawley, C., Heng, S., Vinsonhaler, R., Lauga, E., & Thiffeault, J. L. (n.d.).Publication year
2016Journal title
Physical Review EVolume
94Issue
6AbstractArtificial phoretic particles swim using self-generated gradients in chemical species (self-diffusiophoresis) or charges and currents (self-electrophoresis). These particles can be used to study the physics of collective motion in active matter and might have promising applications in bioengineering. In the case of self-diffusiophoresis, the classical physical model relies on a steady solution of the diffusion equation, from which chemical gradients, phoretic flows, and ultimately the swimming velocity may be derived. Motivated by disk-shaped particles in thin films and under confinement, we examine the extension to two dimensions. Because the two-dimensional diffusion equation lacks a steady state with the correct boundary conditions, Laplace transforms must be used to study the long-time behavior of the problem and determine the swimming velocity. For fixed chemical fluxes on the particle surface, we find that the swimming velocity ultimately always decays logarithmically in time. In the case of finite Péclet numbers, we solve the full advection-diffusion equation numerically and show that this decay can be avoided by the particle moving to regions of unconsumed reactant. Finite advection thus regularizes the two-dimensional phoretic problem.