1, for a multiple linear regression algorithm to the data set which has better explanatory, but we can compare the size of the output coefficients of the characteristic parameters of its data to determine the impact of heavy weights, and further wherein the implicit parameter for expansion and collection, improve the accuracy of the overall training data.
2, mathematical principles and program underlying multiple regression algorithm is written as follows:
Multivariate linear regression algorithm can be written from the entire bottom package according to the above mathematical principle is as follows:
The whole package together multiple linear regression algorithm can directly call.