Premium
Graphics for uncertainty
Author(s) -
Bowman Adrian W.
Publication year - 2019
Publication title -
journal of the royal statistical society: series a (statistics in society)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.103
H-Index - 84
eISSN - 1467-985X
pISSN - 0964-1998
DOI - 10.1111/rssa.12379
Subject(s) - animation , computer science , graphics , representation (politics) , context (archaeology) , statistical graphics , simple (philosophy) , variety (cybernetics) , computer graphics , contingency table , theoretical computer science , computer graphics (images) , human–computer interaction , data mining , artificial intelligence , machine learning , paleontology , philosophy , epistemology , politics , political science , law , biology
Summary Graphical methods such as colour shading and animation, which are widely available, can be very effective in communicating uncertainty. In particular, the idea of a ‘density strip’ provides a conceptually simple representation of a distribution and this is explored in a variety of settings, including a comparison of means, regression and models for contingency tables. Animation is also a very useful device for exploring uncertainty and this is explored particularly in the context of flexible models, expressed in curves and surfaces whose structure is of particular interest. Animation can further provide a helpful mechanism for exploring data in several dimensions. This is explored in the simple but very important setting of spatiotemporal data.