Often (?!) we have to deal with questions like ‘are the groups different?’, ‘on what variables, are the groups most different?’, ‘can one predict which group a person belongs to using such variables?’ In answering such questions, discriminant analysis is quite helpful.
[Read More]
Projection Pursuit
A dimensionality reduction through orthogonal transformation
Projection Pursuit seek to find seeks to find one- and two-dimentional linear projections of multivariate data that are relatevely highly revealing.
[Read More]
Multidimensional Scaling
Multidimensional scaling (MDS) is one of the many methods for dimensionality reduction. Actually, it more a family of techniques rather than a single method, whose goal is to produce a mapping of our N-dimensional data in a lower dimensionality space.
[Read More]
Principal Components Regression
Principal Components Regression (PCR) is a regression technique based on principal component analysis (PCA).
[Read More]
Principal Component Analysis
A dimensionality reduction through orthogonal transformation
Principal components analysis (PCA) is used when a simpler representation is desired
for a set of intercorrelated variables. It is a method of transforming the original
variables into new, uncorrelated variables. The new variables are called the
principal components.
[Read More]
Partial Least Squares Discriminant Analysis
A PLS-R variant for categorical variables
It may happen that you have a dataset with columns $X_1, …, X_n, Y$, and you want to check if there is any correlation between $X_1, …, X_n$ variables (or part of them) and the $Y$ variable. You know that correlation doesn’t mean causation, but sometimes it would be useful...
[Read More]
Sammon mapping
A non-linear mapping for data visualization
Sammon Mapping belongs to the multidimensional scaling algorithms family, as its main goal is to reduce a high dimensionality space to a lower dimensionality space, for visualization purposes. Unlike PCA and other dimensionality scaling algorithms, Sammon Mapping’s aim is not to highlight the most descriptive component to project the original...
[Read More]