Research Interests:

  • Network Analysis
  • (Hard-to-Reach) Population Size Estimation
  • Computational Methods
  • Statistical Consulting
  • Statistical Applications in Social Sciences

Publications:

  1. Christian S. Schmid, Ted H. Chen and Bruce A. Desmarais (2021). Generative Dynamics of Supreme Court Citations: Analysis with a New Statistical Network Model. Accepted for publication at Political Analysis. https://arxiv.org/abs/2101.07197
  2. P.W. Thurner, C.S. Schmid, S.J. Cranmer and G. Kauermann (2018). Network Interdependencies and the Evolution of the International Arms Trade. Journal of Conflict Resolution. https://doi.org/10.1177/0022002718801965
  3. C. S. Schmid and B. A. Desmarais. Exponential random graph models with big networks: Maximum pseudolikelihood estimation and the parametric bootstrap. 2017 IEEE International Conference on Big Data (Big Data), pages 116–121, Dec 2017;  http://ieeexplore.ieee.org/document/8257919/  

Working papers:

  1. Christian S. Schmid, David R. Hunter and Pavel Krivitsky. Improving ERGM Starting Values Using Simulated Annealing. https://arxiv.org/abs/2009.01202
  2. Christian S. Schmid and David R. Hunter. Accounting for Model Misspecification When Using Pseudolikelihood for ERGMs.
  3. Bomin Kim, Aaron Schein, Bruce A. Desmarais, Hanna Wallach and Christian S. Schmid. The Hyperedge Event Model.

R-packages:

  1. Author of cERGM: Fit, Simulate and Diagnose Citation Exponential Random Graph Models 
  2. Contributor of ergm: Fit, Simulate and Diagnose Exponential-Family Models for Networks

Research Assistant:

  1. Summer 2018: Xiaoyue Maggie Niu, Department of Statistics, Pennsylvania State University
  2. 2016 – present: Bruce A. Desmarais, Department of Political Science, Pennsylvania State Univerisity
  3. Summer 2017: Le Bao, Department of Statistics, Pennsylvania State University
  4. 2014 – 2015: Paul W. Thurner, Department of Political Science, Ludwig Maximilians University Munich

Projects:

  1. Estimating the Size of Populations with High Risk for HIV (May 2017 – present)
    Developed new models for Aggregated Relational Data with the goal of estimating the size of hidden and hard-to-reach population groups. Implemented Bayesian hierarchical models using R, Stan and C++.
  2. Inferential Analysis of the Supreme Court Citation Network (August 2017 – present)
    Developed a new statistical model for dynamic citation networks with the goal of understanding what drives the use of opinions as precedents through the study of Supreme Court case citation patterns. Currently developing techniques for more computationally efficient parameter estimation.
  3. Credit Card Fraud Detection (January 2018 – May 2018)
    Used under and over-sampling to alleviate unbalanced data and used Naive Bayes, Logistic Regression, Discriminant Analysis, Support Vector Machines, Random Forests and Neural Networks to predict fraudulent transactions.
  4. Maximum Pseudolikelihood and Parametric Bootstrap for ERGM (May 2016 – August 2017)
    Developed new estimation techniques that allow parallelization of the computationally expensive estimation of Exponential Random Graph Models (ERGM). Applied statistical network theory to Political Science data sets in R.
  5. Modeling the International Arms Trade Network (October 2014 – October 2016)
    Modeled the International Arms Trade network for Major Conventional Weapons using temporal exponential random graph models and non-parametric bootstrap approaches. Evaluated the predictive power of these models based on ROC and precision-recall curves.