With both exploratory factor analysis [ 23] and confirmatory factor analysis [ 24] we are interested in explaining the variability in the observable variables (e.g. symptoms, questionnaire data) by modelling common latent factors (e.g. depression severity).
Partial least-squares discriminant analysis (PLS-DA) finds a linear regression model by projecting the predicted variables and the observable variables to a new space.
Partial least squares ("PLS", [ 35 ]) can be seen as a technique related to Principal Components Regression but which fits a linear regression model by projecting the predicted variables and the observable variables to a new space where the relation between the variables can be better visualized.
In contrast to previous applications (Beal et al., 2002; Sohn and Xing, 2007), a posteriori inference of the state sequence is furthermore not necessary, since the states (peptides) are already assigned to the observable variables (tandem mass spectra) by means of the corresponding peptide-spectrum matches.
A Bayesian Network (BN) is the joint probability distribution of a set of variables: based on the state of the observable variables and a priori probabilities represented by the edge in the relations between variables, the a posteriori probabilities of the unknown states are evaluated.
The former describes the relation between the obvious variables and the potential variables, indicating the observable variables of the potential variables [10].
In a CRF, the observable variables (X = { X i}) and unobservable variables (Y = { Y i}) are treated separately, with the unobservables globally conditioned on the observables.
Subsequent PLSR analysis finds latent variables of a linear model that describes the predicted variable (protein abundance) in terms of the observable variables.
Partial least-squares (PLS) regression is a statistical method that develops a linear regression model by projecting the predicted variable (% parasitemia) and the observable variable (spectra) onto a new multidimensional space.
The latter involve both the spatial dependence structure of the variable of interest and its relationship with the observable variable.
