Correction for multicollinearity between the explanatory variables to estimation by using the principal component method

Abstract EN

In the most of applications of regessforr, no explanatory orthogonal will exist, but it is connected too strongly to the extent that the results would be far from being exquisite.

So it's too difficult to expect the effects upon the individual variables within the range of regression equation.

Also the values estimated here concerning the factors could be slight in data.

The non-orthogonally is said to be the problem of multicolinearity going side by side with the factors of unstable,estirnated regression.

This case, however, comes out from the strong linear relation between explanatory variants.

To solve this problem, a method of the principal components is used ; that which depends upon the fact that each linear type might be reformulated as to the group of the orthogonal.

explanatory variables : these in turn can be obtained as linear structures for the orthogonal (basic) explanatory variables through the Barr let norm, as a test formula, as a way to know whether the roots possess a sufficient qualify for a linear relation As for the practical side, or applied of this research concerning the special data of the consumption of the individual in USA as dependent variable and wage income.

Non-wage-non farm income, farm income are explanatory variables.

The characters root indicated to the collinearity ; the result is that the four variables can be treated as two factors only.

a statistical programmer is here used that is (SPSS, Minitab) for the analysis of the data.