Selected Publications

We investigate the complications and opportunities when drawing causal inference from spatial observational data. We introduce causal diagrams that allow us to investigate the impact of spatial confounders, interference, and the inherent spatial structure in the exposure variable, and we illustrate that causal inference with spatial data has crucial differences to counterparts with independent observations. We then propose an approach that mitigates bias from unmeasured spatial confounding and incorporates interference within one framework.
arXiv

In cluster randomized experiments with selection bias due to recruitment, data are often only available on those that were recruited. Based on a principal stratification framework, we show that causal effects on the overall population are identifiable based on the recruited sample only.
PDF arXiv R package

Estimating the parameters of a temporal, spatio-temporal, or mutually-exciting Hawkes process based on data that are available in aggregated form by time, space, or both.
Code arXiv

A framework for causal inference with bipartite interference from observational, time series data with a random bipartite network.
PDF arXiv
JASA

We propose a latent factor interaction model for networks measured with error, and a variable importance metric for latent models. We use the model to address the geographic and taxonomic bias of ecological studies of species’ interactions, and identify the important bird and plant covariates for forming and detecting interactions.
Dataset Slides Video Journal arXiv R package Analysis
JRSS-B

Evaluating the effect of a point pattern treatment on a point pattern outcome measured over time.
PDF arXiv Journal R package
Biometrics

Causal inference with interference for realistic treatment allocation programs. Evaluating the comparitive effectiveness of power plant emission reduction strategies for reducing ambient ozone concentrations.
PDF Dataset R package Journal Slides
Biostatistics

Incorporating geographical distance in a propensity score matching approach to account for unmeasured confounding by spatial variables.
PDF Dataset R package Journal Slides

Publications

We investigate the complications and opportunities when drawing causal inference from spatial observational data. We introduce causal …
arXiv

In cluster randomized experiments with selection bias due to recruitment, data are often only available on those that were recruited. …
PDF arXiv R package

Estimating the parameters of a temporal, spatio-temporal, or mutually-exciting Hawkes process based on data that are available in …
Code arXiv

A framework for causal inference with bipartite interference from observational, time series data with a random bipartite network.
PDF arXiv

Statistical software manuscript introducing the geocausal R package for spatio-temporal causal inference.
OSF CRAN

We propose a latent factor interaction model for networks measured with error, and a variable importance metric for latent models. We …
JASA
Dataset Slides Video Journal arXiv R package Analysis

Estimating the causal effect of the Hospital Readmissions Reduction on program on patient readmission and mortality rates.
HSORM
arXiv Journal

Evaluating the effect of a point pattern treatment on a point pattern outcome measured over time.
JRSS-B
PDF arXiv Journal R package

Biometrics
Journal

We introduce a framework for estimating causal effects of binary and continuous treatments in high dimensions. We show how posterior …
Biometrics
R package arXiv Journal

Clinical Trials
arXiv Journal

Estimation of tensor regression coefficients on a soft version of the PARAFAC approximation.
JMLR
arXiv Journal Slides

Causal inference for subgroup analysis and the Connect-S plot.
Statistics in Medicine
Code arXiv Journal

Extending causal inference in the presence of interference to bipartite settings where the interventional units are different from the …
Statistical Science
Code Arxiv Journal

Addressing confounding bias from unmeasured spatial variables using mixed models.
AOAS
Code arXiv Slides Journal

BA
PDF

Causal exposure-response curve estimation with local confonding adjustment. Different variables confound the exposure-response …
AOAS
R package arXiv Journal

A review of the literature on the health effects of exposure to low levels of particulate matter.
Current Environmental Health Reports
Journal

JASA
PDF Journal

Causal inference with interference for realistic treatment allocation programs. Evaluating the comparitive effectiveness of power plant …
Biometrics
PDF Dataset R package Journal Slides

Incorporating geographical distance in a propensity score matching approach to account for unmeasured confounding by spatial variables.
Biostatistics
PDF Dataset R package Journal Slides