Any statistical technique analysing the relationship between more than two variables; in other words an analysis which looks at the simultaneous and combined effects of a number of variables. Simplifications have to be made; for example, the interrelations between many variables have to be reduced to data on the correlations between each pair.
A frequently used technique in MVA is the use of the multiple linear regression equation:
Yc = a + b1 X1 + b2X2where Yc is the estimated value of the dependent variable, a is the Y intercept, X1, is the value of the first independent variable, X2 the value of the second independent variable, b1 the slope associated with X1, and b2 the slope associated with X2.
MVA methods are commonly used to reduce a large number of intercorrelated variables to a much smaller group, while preserving as much as possible of the original variation, yet having properties such as independence. Such methods include factor analysis and principal components analysis.
