In certain applications, the quality of a process is characterized by the functional relationship between a response variable and one or more explanatory variables. At each sampling stage one observes a set of data points of these variables that can be represented by a curve (or profile) and profile monitoring done for checking the stability of this relationship over time by observing profile data. In some applications profiles can be described reasonably well by a linear regression model but this assumption may not be true in many cases and there more flexible models are to be considered. This article focuses on nonparametric profile monitoring when within-profile data are correlated. (39 refs.)
展开▼