Network Data Analysis
We are interested in the topological structures of various networks, such as the communities in social networks, interactions of different regions of the brain, coauthorship and citations of scientists. Our research topics span from statistical methodology, computational algorithms to applications in complex systems in sciences and business. If you see the potential for collaboration on complex data sets, please do not hesitate to contact us. All our research projects are motivated by real data.
Statistician Coauthorship and Citation Networks
We have collected a data set for coauthorship and citations between statisticians. The citation network of 2654 authors is shown below. We have identified many interesting communities such as "Spatial Statistics", "Large-Scale Multiple Testing", "Variable Selection", "Dimensional Reduction", "Bayes", "Quantile Regression", and "Theoretical Machine Learning", etc. Our paper will appear in the Annals of Applied Statistics with discussion, and be presented in the AOAS Lecture at the Joint Statistical Meetings (JSM) 2016 in Chicago. Please see more details in the paper http://arxiv.org/abs/1410.2840.
The human brain is a complex network of neurons linking physical structure to function. But these links are very challenging to find (the BRAIN Initiative). In the early 20th century German neurologist Dr. Korbinian Brodmann defined 52 regions (see the picture below) of the cerebral cortex of humans and monkeys that appeared to have different cellular morphology and organization. Over the past century clinical findings and neurophysiological studies have shown that these microstructural differences correlate well with cortical function specialization.
The spread of infectious diseases depends on the complex network of contacts. Quantifying the effect of the social networks can help with the control of infectious diseases, which is extremely important for public health. See the review paper.
Savage's approach to research via Mosteller(Hamada and Sitter 2004):
- As soon as a problem is stated, start right away to solve it. Use simple examples.
- Keep starting from first principles, explaining again and again what you are trying to do.
- Believe that this problem can be solved and that you will enjoy working it out.
- Don't be hampered by the original problem statement. Try other problems in its neighborhood; maybe there's a better problem than yours.
- Work an hour or so on it frequently.
- Talk about it; explain it to people.