Contents
Is there a relationship between LDA and PLS-DA?
There is an interesting relationship between LDA and PLS-DA 2.: PLS-DA using the full PLS model (i.e. all latent variables) produces the same predictions as LDA. OTOH, PLS-DA with only one latent variable produces the same predictions as a Euklidean distance classifier (EDC; i.e. assign the class whose mean is closest).
How are selected variables listed in PLS-DA?
Selected variables are listed in the selectVar function: and can be visualised in plotLoadings with the arguments contrib = ‘max’ that is going to assign to each variable bar the sample group colour for which the mean ( method = ‘mean’) is maximum. See example (plotLoadings) for other options (e.g. min, median)
What are the components of a PLS-DA model?
A set of components, also called latent variables. There are as many components as the chosen dimension of the PLS-DA model. A set of loading vectors, which are coefficients assigned to each variable to define each component. These coefficients indicate the importance of each variable in PLS-DA.
Which is the rule of thumb for PLS-DA?
1 – The number of components to retain ncomp. The rule of thumb is usually K −1 K − 1 where K K is the number of classes, but it is worth testing a few extra components. 2 – The number of variables keepX to select on each component for sparse PLS-DA, 3 – The prediction distance to evaluate the classification and prediction performance of PLS-DA.
Which is better PLS or OPLS or PLS-DA?
In this way any misguiding claim, like OPLS is better tha PLS-DA, PLS-DA is better than PLS can be validated with a set of independent methods, like a multi-class ANOVA or a simple feature selection process (for biomarker finding) with PCA (for visualization). PLS-DA loadings plot (left) and PLS-DA scores plot (right).
How is PLS-DA based on PLS regression?
PLS-DA is based on PLS regression (PLS-R) with the Y variable generated from experimental group membership, mapped into a linear space. In a 2-group experiment this can be as simple as 0 and 1.
When to use partial least squares ( PLS ) DA?
Partial least squares-discriminant analysis (PLS-DA) for classification of high-dimensional (HD) data: a review of contemporary practice strategies and knowledge gaps