Redundancy Analysis (RDA) in R - Part I скачать в хорошем качестве

Redundancy Analysis (RDA) in R - Part I

Redundancy Analysis RDA is the multivariate (meaning multi response) technique analogue of regression. The method uses a mix of linear regression and principal components analysis (PCA). Conceptually, RDA is a multivariate (meaning multiresponse) multiple linear regression followed by a PCA of the table of fitted values. Pre-analysis 1. If your response variables are not dimensionally homogeneous (i.e. if they have different base units of measurement), you may center them on their means or standardize them. 2. Ensure the number of explanatory variables is less than the number of objects (sites, samples, observations etc.) in your data matrices. 3. If your explanatory variables are not dimensionally homogeneous (e.g. have different physical units), center them on their means and standardize them. Standardization allows direct comparison of regression coefficients, which may have different scales otherwise. 4.If the relationships are markedly non-linear, apply transformations to linearize the relationships and reduce the effect of outliers. 5. If you wish to represent non-Euclidean relationships between objects in an RDA ordination, you should apply an ecologically-motivated transformation