Top Page | Upper Page | Contents | About This Site | JAPANESE

Multi-Regression Analysis

Multi-Regression Analysis is a kind of Regression Analysis . It uses more than one X values.

Multi-Regression Analysis is the basic of Multi-Variable Analysis and Data Mining.

Formulation

In multi regression analysis, the formula is
Y = A1 * X1 + A2 * X2 + EEE + An * Xn + B

How to understand the conclusion

Coefficient Evaluation

We can judge the goodness of fitting with coefficient of determination.

In multi regression analysis, we can see the effect of each Xi by Ai. But before this analysis, we need Normalization.

By the way, in single regression analysis, when we take Normalization,
Ai = correlation coefficient. So the analysis of Ai is equal to the analysis of correlation coefficient.

For Category Data

When X is category data Category Data, regression analysis is Broadly defined Quantification theory 1.

Path Analysis

Actual data has network structure. In such network, X is Y of other X. To evaluate network structure, there is Path Analysis . Path Analysis is not only an application of multi Regression Analysis.




NEXT Multi vs Single