Excellent thread on the containment trap 👇
Ebola is back in the DRC. We know epidemics and conflict tend to coincide — but how, why, and when does an epidemic actually intensify civil conflict? My job market paper offers an overarching framework. 🧵
Mitsuru Mukaigawara, MD, MPP
@mitsurumu.bsky.social
Physician-political scientist and applied statistician @harvard.edu. Radcliffe Fellow @rad-institute.bsky.social. Website: mitsurumukaigawara.com ProjectGEOCAUSAL: geocausal.org
Excellent thread on the containment trap 👇
Ebola is back in the DRC. We know epidemics and conflict tend to coincide — but how, why, and when does an epidemic actually intensify civil conflict? My job market paper offers an overarching framework. 🧵
Ebola is back in the DRC. We know epidemics and conflict tend to coincide — but how, why, and when does an epidemic actually intensify civil conflict? My job market paper offers an overarching framework. 🧵
I’m delighted to share that I’ll be joining the Radcliffe Institute’s 27th class of fellows this fall, where I’ll work on my book project on epidemics, conflict, and international security.
Who’s coming to Harvard Radcliffe Institute in the fall? Meet the scientists, writers, scholars, public intellectuals, and artists who will explore the frontiers of knowledge and practice. www.radcliffe.harvard.edu/news-and-ideas/harvard-radcliffe-institute-announces-2026-2027-fellows
Our new ProjectGEOCAUSAL website is live! We’ll be posting the latest updates on cutting-edge applications of spatiotemporal causal inference to address the world’s most pressing problems. Check it out: geocausal.org
We propose a spatiotemporal causal inference framework that fully leverages microlevel, granular data. ATE, heterogeneity, and mediation — all in one framework. Now with updated results and visualizations!
🚨New paper🚨: Are you interested in key debates about civil war and counterinsurgency? Like questions about causal inference but wish you had a better method for dealing with spillover and carryover effects? Then check out our new paper: arxiv.org/abs/2504.03464
New working paper: “Survey Estimates of Wartime Mortality,” with Gary King, available at gking.harvard.edu/sibs. We provide the first formal proofs of the statistical properties of existing mortality estimators, along with empirical illustrations, to develop intuitions that guide best practices.
Excited to present our poster on spatiotemporal causal inference at #PolMeth 2025. Looking forward to seeing many of you there! Paper: arxiv.org/abs/2504.03464 Package: github.com/mmukaigawara...
Interested in causal inference using high-frequency, fine-grained geospatial data? Check out our 2025 PolMeth poster on spatial-temporal causal inference, designed by the wildly talented @mitsurumu.bsky.social. We examine the effects of US airstrikes and civilian harm on insurgent attacks in Iraq
Interested in causal inference using high-frequency, fine-grained geospatial data? Check out our 2025 PolMeth poster on spatial-temporal causal inference, designed by the wildly talented @mitsurumu.bsky.social. We examine the effects of US airstrikes and civilian harm on insurgent attacks in Iraq
New paper alert (hey, I can't doom scroll all the time): This one's on doing causal inference with "microlevel data" where we suspect that the treatment has spatial spillover & temporal carryover effects. We illustrate our new approach + package w/ application to US counterinsurgency efforts in Iraq
Spatiotemporal causal inference with arbitrary spillover and carryover effects
Micro-level data with granular spatial and temporal information are becoming increasingly available to social scientists. Most researchers aggregate such data into a convenient panel data format and a...
arxiv.org
How can we identify causal effects using micro-level data? Our new framework estimates ATEs, probes causal mechanisms, and uncovers heterogeneity—all in one. We illustrate it with an analysis of airstrikes and insurgent attacks in Iraq. arxiv.org/abs/2504.03464
Spatiotemporal causal inference with arbitrary spillover and carryover effects
Micro-level data with granular spatial and temporal information are becoming increasingly available to social scientists. Most researchers aggregate such data into a convenient panel data format and a...
arxiv.org
Here's one example of how we used geocausal to estimate the effects of different distributions of US airstrikes in Iraq on insurgent attacks (open access): academic.oup.com/jrsssb/artic...
Causal Inference with Spatio-Temporal Data: Estimating the Effects of Airstrikes on Insurgent Violence in Iraq
Abstract. Many causal processes have spatial and temporal dimensions. Yet the classic causal inference framework is not directly applicable when the treatm
academic.oup.com
Many of us work with high-frequency spatial & temporal data that can frustrate causal inference. That's why we created geocausal, a new R package that helps you identify counterfactuals for user-specified windows even in the face of spillover & carryover effects. Check it out:
Want to do causal inference using high-frequency geospatial data with temporal carryover or spatial spillover effects? We've created a new R package, geocausal, that lets you estimate counterfactuals at user-specified intervals for different distributions of a treatment github.com/mmukaigawara...
GitHub - mmukaigawara/geocausal: Causal inference with spatio-temporal data in R
Causal inference with spatio-temporal data in R. Contribute to mmukaigawara/geocausal development by creating an account on GitHub.
github.com
New in AJPS with @carlynwayne.bsky.social @mitsurumu.bsky.social @profmholmes.bsky.social: how do group dynamics affect assessments of resolve and costly signals? #polisky onlinelibrary.wiley.com/doi/10.1111/...
Interested in doing causal inference with spatio-temporal data? We've got a new R package, geocausal, that allows you to estimate causal effects + counterfactuals for super fine grained data over user-specified spatial + temporal windows. Download it here. (Paper TK soon). polisky
GitHub - mmukaigawara/geocausal: Causal inference with spatio-temporal data in R
Causal inference with spatio-temporal data in R. Contribute to mmukaigawara/geocausal development by creating an account on GitHub.
github.com