Yuefeng Wu

Yuefeng Wu
Email (1892-03-30)wuyue (at) umsl (dot) edu
Phone (1892-03-30)314-516-6348
  (1892-03-30)Department of Mathematics and Computer Science
Address (1892-03-30)320 ESH
  (1892-03-30)One University Blvd.,
  (1892-03-30)Saint Louis, MO 63121

I am an assistant professor in the Department of Mathematics & Computer Science at the University of Missouri at Saint Louis. My doctoral degree in Statistics was obtained from North Carolina State University. In Cornell University, two years had been spent for my post doctoral training, and in University of California Santa Cruz, I had worked as a visiting assistant professor for another two years.

I have a number of research areas and active projects, such as Bayesian (nonparametric) methods, disparity-based inference, pattern recognition, social network data analysis, machine learning in big data, and inference for dynamical systems and differential equations. Part of my research was supported by grants from NASA.



Research Interests

My research focusses on a number of issues within these five fields:

  • Inference on dynamical systems and differential equations: I focus on giving statistical inferences on both the observational error and the stochastical disturbance of the dynamical systems that can be modeled by ordinary differential equations with stochastical disturbance terms.
  • Bayesian (nonparametric) methods: I worked on Dirichlet process kernel mixture type of priors. I am also working on developping some MCMC algorithm for Bayesian methods, parametric or nonparametric.
  • Machine learning and big data: In particular, I am interested in pattern recognition, syntactic methods, and parallel computing for big data. All these works relate to the NASA projects, whose goal is to design and test the next generation air traffic control system.
  • Disparity-based inference: These involve first estimating a non-parametric version of the model and then comparing this estimate to a parametric description. The extra smoothness that you gain from the non-parametric estimate allows you to use a comparison metric -- Hellinger distance is the best know of these -- that makes your parameter estimates insensitive to outlying data points without giving up statistical precision. 
  • Social network data analysis: A series of models for considering both the attribution value of each nodes and the adjacency matrix among them simutanously to model the random graph of the social network are under developping and testing.


  • NASA UARC Identifying Major Contributions to Trajectory Prediction Errors 444065- 26816
  • NASA UARC Air Traffic Management Fundamental Research Task 444065-26817
  • FY2013 Seedling Awards for Using Historical Data to Automatically Identify Air-Traffic Controller Behavior.  WBS Element 694478.
  • FY2014 NASA UARC Task 057


Prospective Publications:

These are a range of paper ideas, some of them more likely to turn into papers than others, some of them larger projects than others, that I think worthwhile. Anybody who is interested in them, has relevant data, or knows of authors that have beaten me to it is highly encouraged to contact me. Many of these are also potential graduate student projects.
  • Hadoop for Syntactic Pattern Recognition
  • Estimation of Mutual Information based on Bayesian nonparametric density estimation
  • Consistency of Chinese Restaurent Process for Random Blocking Model
  • From empirical mode decomposition to Synchro-squeezing transform for Big Data in ATC identification
  • Inference on noisily observed dynamic systems with stochastic disturbances
  • Models for Social Network with Attribution values


The above animation shows one day's air traffic in the airspace of U.S. One of my current working projects is to identitying the unusual maneuvering of these filights due to the Air Traffic Controllers' requests. One day's flights generate data about 60 Gigabytes, and we are aiming to analyze the whole data in 10 years.


  1. Jump up BioStatistics (Fall 2011, Winter 2012)
  2. Generalized Linear Model (focused on Bayesian approach) (Spring 2012)
  3. Linear Modelump up to: (Spring 2013)
  4. Jump up Teaching and Research on AMS (Fall 2012)
  5. Applied Statistics (Fall, 2013, 2014, Spring 2015)
  6. Mathematical Statistics I (Fall, 2013, 2014)
  7. Mathematical Statistics II (Spring 2014, Spring 2015)

Statistical Consulting

In addition to the above, part of my job involves statistical consulting. I have worked on a wide range of applied problems since 2011. Anybody who has data analysis related problems is highly encouraged to contact me.

Jump up

 In addition to the areas above, part of my job involves statistical consulting to the Cornell community and I have been involved in a wide range of applied problems. I have also worked on Item Response Theory and its applications to analyzing web browsing behavior, educational testing and medical diagnostics.