Multiresolution network models
Duration: 34 mins 52 secs
Share this media item:
Embed this media item:
Embed this media item:
About this item
Description: |
McCormick, T (University of Washington)
Tuesday 26th July 2016 - 13:30 to 14:00 |
---|
Created: | 2016-07-28 15:16 |
---|---|
Collection: | Theoretical Foundations for Statistical Network Analysis |
Publisher: | Isaac Newton Institute |
Copyright: | McCormick, T |
Language: | eng (English) |
Abstract: | Social networks exhibit two key topological features: global sparsity and local density. That is, overall the propensity for interaction between any two randomly selected actors is infinitesimal, but for any given individual there is massive variability in the propensity to interact with others in the network. Further, the relevant scientific questions typically differ depending on the scale of analysis. In this talk, we propose a class of multiresolution statistical models that model network structures on multiple scales to enable inference about relevant population-level parameters. We capture global graph structure using a mixture over projective models that capture local graph structures. This approach is advantageous as it allows us to differentially invest modeling effort within subgraphs of high density, while maintaining a parsimonious structure between such subgraphs. We illustrate the utility of our method using simulation and data on household relations from Karnataka, India. This is joint work with Bailey Fosdick (CSU) and Ted Westling (UW). |
---|
Available Formats
Format | Quality | Bitrate | Size | |||
---|---|---|---|---|---|---|
MPEG-4 Video | 640x360 | 1.94 Mbits/sec | 507.70 MB | View | Download | |
WebM | 640x360 | 633.58 kbits/sec | 161.80 MB | View | Download | |
iPod Video | 480x270 | 522.21 kbits/sec | 133.36 MB | View | Download | |
MP3 | 44100 Hz | 249.76 kbits/sec | 63.84 MB | Listen | Download | |
Auto * | (Allows browser to choose a format it supports) |