Further Information
Wayne Oldford:
Selected Papers:
Visualizing Dependence in High-Dimensional Data: An Application to S&P 500 Constituent Data (Marius Hofert, Wayne Oldford) In press 2018 Econometrics and Statistics: www.sciencedirect.com/science/article/pii/S245230621730031X
Euclidean distance matrix completion and point configurations from the minimal spanning tree (Adam Rahman, Wayne Oldford) SIAM Journal on Optimization, 2018: epubs.siam.org/doi/10.1137/16M1092350
A framework for measuring dependence between random vectors (Marius Hofert, Wayne Oldford, Avinash Prasad, Mu Zhu) (submitted 2018 to J. Multivariate Analysis) arxiv.org/abs/1801.03596
YouTube channel:
Lectures from Fall 2017 Data Visualization course @Waterloo:
www.youtube.com/playlist
Most relevant to the talk are those numbered 19a to 23 inclusive. Here are links to a few of those:
19c: Non-Cartesian: Interactive parallel coordinates: youtu.be/g9zk66dmCRE
19e: Non-Cartesian: space filling glyphs for high-dimensional data: youtu.be/MVfBRE0pTIA
20a: Projections: navigation graphs youtu.be/MVfBRE0pTIA
20b: Projections – navigation graphs and zenplots youtu.be/QU2sOe0dm-c
20c: Projections – interesting projections, scagnostics youtu.be/ZmB3MeHprGg
21b: Reducing Dimensionality – Principal components – Frey faces example youtu.be/8uc3c_aYqOc
Software:
Much of the software used is the loon package in R. Those interested in exploring it themselves might visit some online documentation and examples at
Loon: An interactive statistical visualization toolkit: waddella.github.io/loon/ (from CRAN, in R: install-packages(“loon”) )