We provide a smooth introduction to reduced-rank analysis via singular-value decomposition, and show how it can be used to monitor images of etched silicon wafers. An industrial case study is discussed.

Circular data are measured in angles or directions, which occur a wide variety of scientific fields. An often investigated hypothesis is that circular uniformity, isotropy. Frequentist methods for assessing the uniformity null exist, but do not allow user faced with an insignificant result to distinguish lack power from support hypothesis. Bayesian tests, solve this issue and several others, de...