Stat 319: Literature of Statistics (networks and graphs)


The study of network/graph based data has a very long history. The past ten years or so have seen an explosion in the number of contexts and the size of data sets of this kind.

Many interesting new statistical problems have appeared. Students in this class will read and present papers on these topics.

Example topics

  1. Finding and testing communities in graphs.
  2. Studying time trends in graphs that evolve.
  3. Labeling nodes of a graph based on a small sample of labels.
  4. Predicting edges that will appear or disappear.
  5. Models for random graphs.
  6. Consequences of graph sampling methods.
  7. Surveying and curating graph data set repositories.
  8. Surveying and curating R/Matlab/python/etc code for handling and displaying graphs.
Each of these topics has a literature worth surveying and each is undergoing rapid development. Student may also try 'stalker-mode' research where they read, survey and extend the key papers of a prominent networks researcher or research group.

There is no conflict in taking this course and other network related courses such as Professor Montanari's Stat 375. The only constraint is that the work done for this course must not duplicate work done for another.


Art Owen
Sequoia Hall 130
My userid is owenbuzzard on (remember to remove the carrion eaters)
Office hour: Tuesday 11:00-12:00

Organizational meeting

Monday January 3 1:15 Fishbowl. Regular meetings Tuesdays 1:15 420-050. We will not meet Tuesday January 4.


Here are some good starting points for topics, among many,


Date Speaker Topic
Jan 18 Michael Koenig Econometric time series of graphs
Myunghwan Kim Kronecker random graphs
Rense Corten A large social data network
Jan 25 Sumit Mukherjee Exponential familes of graphs
Hao Chen Communities and more
Sarah Koo Repositories of graph data
Feb 1 Alexandra Chouldechova Link prediction | see figures 1,2,3,5,8,9,11
Michael Lim Information diffusion through graphs
Feb 8 Luo Lu Snijders paper and MCMC
Sarah Koo Software for graph computations
Feb 15 Myunghwan Kim Fitting models like MAGFit to networks
Feb 22 Hao Chen Community detection in protein networks
Alexandra Chouldechova Link prediction in relational data
Mar 1 Michael Lim Detecting emerging trends
Sumit Mukherjee Exponential random graph models
Mar 8 Luo Lu Large deviations for Erdos-Renyi


Evaluation is based on presentations, a write up, and class participation.

Be sure to give Axess a working email address: