The exploitation of chemometric methods in the analysis of spectroscopic data: Application to olive oils

Multivariate analysis is the term used to describe the analysis of data where numerous observations or variables are obtained for each object studied (Afifi and Clark, 1996). The identity or value of any one object sample will be reflected in some or all of the variables measured to a greater or lesser degree. With spectroscopic data it is not generally possible to identify or quantify an object from one variable (Mark, 1991). This may, however, be achieved by disentangling the complicated interrelationships between a number of the variables by means of multivariate statistical methods (Martens and Næs, 1989).