David Carlson

Data Science and ML/AI

~ Preview ~

Welcome! I am an accomplished data scientist and ML/AI engineer. I have a PhD in Political Methodology from Washington University in St. Louis. After my studies, I was a computational social sciences and statistical methodology assistant professor at Koc University in Istanbul. I have since entered the industry, and continue to develop and advance my research in these areas, often with social science applications. I love working with data, modeling, and constantly learning. Please, do not hesitate to reach out!

Peer-Reviewed Publications

Boston, Joshua, David Carlson, JB Duck-Mayr, and G Sasso. Political Competition and Judicial Independence: How Courts Fill the Void When Legislatures are Ineffective. Journal of Law and Courts (2024): 12, 1.
Marcus, J., Carlson, D., Ceylan, S., Ergin, C. Generation Me: A generational explanation for intra-national cultural convergence and divergence in self vs. collective cultural orientations. Journal of World Business (2021): 101269.
- Supplementary Information
Carlson, David and Jeffrey Ziegler. The Role of Democratic Governance and Indirect Expropriation in International Investment Treaty Violations. Accepted at Uluslararasi Iliskiler / International Relations Journal (2021).
- Supplementary Information
Aksoy, Deniz, and David Carlson. Electoral Competition and Militant Groups Targeting Strategies. Accepted at Journal of Peace Research (2021).
- Supplementary Information
Hadzic, Dino, David Carlson, and Margit Tavits. How Exposure to Violence Affects Ethnic Voting. British Journal of Political Science (2020): 50 (1), 345362.
- Supplementary Information
Carlson, David. Modeling Related Processes with an Excess of Zeros. Political Science Research and Methods (2019): 7 (4), 889902.
- Supplementary Information
Carlson, David and Jacob M. Montgomery. A pairwise comparison framework for fast, flexible, and reliable human coding of political texts. American Political Science Review 111, no. 4 (2017): 835843.
- Supplementary Information

Working Papers

Carlson, David. Modeling Without Conditional Independence: Gaussian Process Regression for Time-Series Cross-Sectional Analyses.
Carlson, David, and Abdulhakim Özcan. Estimating Causal Orderings and Relationships with Causal Gaussian Process Regression.
Carlson, David, and Devin P. Brown. Estimating a Counter-Factual with Uncertainty Through Gaussian Process Projection.

Open-Source Software

R package GPP to project a counter-factual with uncertainty through Gaussian Process Projection (also available on CRAN).
Instructions and JavaScript code for SentimentIt API and R package.
R package weightTAPSPACK to subset, weight, and impute data missing from attrition for The American Panel Survey (also available on CRAN).
JAGS code for Zero-Inflated Multivariate Ordered Probit model, presented in Modeling Related Processes with an Excess of Zeros.
Stan code for Gaussian process regression for time-series, cross-sectional data.