Comparative study on the use of statistical methods for the distribution of autocorrelation to any influence factor

Three methods reported in the literature are subject to comparative analysis in the present paper: 1. Classic method [1,5]; 2. Merce E., Merce C.C. Method[2,3]; 3. Merce E., at al Method[4]; It is shown that in the case of the first two methods mentioned above, the attempts to distribute interactions on influence factors have as a prerequisite the determination of the simple correlation coefficients and of the partial correlation coefficients, the methods being of this particularly laborious nature. With obvious computational facilities, compared to the first two methods, the authors propose the use of a new method based on the principle of proportional distribution of autocorrelation with the coefficients of simple determination, and the following five steps are being performed: 1) Calculation of multiple correlation coefficient and simple correlation coefficients using the Regression function of the Microsoft Excel Data Analysis component; 2) The recording of the multiple correlation coefficient and the simple correlation coefficients in the Excel table used for this purpose; 3) Calculating the coefficients of the simple determination and the multiplication factor; 4) Sum of coefficients of simple determination; 5) Calculating the proportions of simple determinations, considering their sum equal to 100; 6) Determination of the influence of each factor as a product between multiple determinations and the proportion of simple determinations. Note that the last four steps in the Excel work table are generated instantly after the first two steps.