(PCA) A procedure that attempts to identify patterns in a large dataset of variable character states.
Visual and numerical analysis of collinearity among variables. Allows for the mapping of high-dimensional data to a lower dimension for visualisation, analysis and modelling.
A method of analyzing multivariate data in order to express their variation in a minimum number of principal components or linear combinations of the original, partially correlated variables.
A statistical procedure for reducing the dimensionality of a data set by approximating the given variables as linear combinations of factors than are uncorrelated with each other. A large number of input variables will usually yield a much smaller set of factors. Most statistical packages provide principal components analysis.
An ordination technique for analyzing data from several variables, such as allelic frequencies or morphological data. The method finds linear trends (principal components) through the clouds of sample points in multidimensional space. These principal components account for the greatest amount of variation present in the data. The residual variance is removed from the data with the calculation of each successive principal component.
Statistics: Principal component analysis (PCA) is a multivariate statistical procedure that was developed by Hotelling in 1933. It is a process where a large set of correlated variables is reduced to a smaller set of principal components or latent factors. The reduction is accomplished by transforming the original variables to a new set of uncorrelated linear combinations of the original variables.
The analysis of covariance in a multiple data set so that the data can be projected as additive combinations onto new axes (principal components). The first principal component is defined to be aligned with the direction of maximum variance in the original image data.