Collinearity or multicollinearity refers to the existence of strong
linear relationships among the predictor variables, which means
that one predictor variable can be near-linearly predicted from the
others. When there is no linear relationship among predictor
variables at all, they are said to be orthogonal. The lack of
orthogonality among the predictor variables usually is not strong
enough to affect the analysis or the ability of the entire number of
predictors to predict the response variables. In other words, the
lack of orthogonality does not diminish the usefulness of the
model, at least within the sample data used tofind the regression
coefficients. However, this condition can produce ambiguous
results that are associated with unstable estimated regression
coefficients and affects the calculations associated to individual
predictors. Instability in the estimated coefficients can be indicated by large changes in the estimated regression coefficients
when a variable is added or deleted, or when a data point is
altered or dropped. When dealing with collinearity, the principal
component analysis (PCA) method is one of the most common
ways to reduce collinearity[25]. The PCA, contrary to grouping
methods such as cluster analysis, is a one-sample technique that
uses orthogonal transformation to obtain a set of values of linearly
uncorrelated variables called principal components, which can be
equal to or less than the original number of predictor variables