Barycentric Discriminant Analysis

Barycenter The mean of the observations from a given category (also called center of gravity, center of mass, mean vector, or centroid) Confidence interval An interval encompassing a given proportion (e.g., 95%) of an estimate of a parameter (e.g., a mean) Discriminant analysis A technique whose goal is to assign observations to some predetermined categories Discriminant factor scores A linear combination of the variables of a data matrix. Used to assign observations to categories Design matrix (aka group matrix) In a group matrix, the rows represent observations and the columns represent a set of exclusive groups (i.e., an observation belongs to one and only one group). A value of 1 at the intersection of a row and a column indicates that the observation represented by the row belongs to the group represented by the column. Avalue of 0 at the intersection of a row and a column indicates that the observation represented by the row does not belong to the group represented by the column Estimation bias The difference between the computed value of a barycenter and the mean of the bootstrapped estimates of this barycenter Fixed effect model Analysis in which the observations that are predicted were used to compute the predictive model