Abstract
The problem of structural-parametric synthesis of multiple linear regression models on the input variables in conditions of partial multicollinearity is solved in the work. The partial multicollinearity phenomenon of input arguments conduces to a variance estimates increase of the regression model parameters, and makes difficult to explain the impact of input variables to the dependent variable. Substantial multicollinearity conduces to impossibility of output estimation model. Classical stepwise procedures of the factors selection do not solve the problem of multicollinearity. Thus, the design of structural-parametric multiple regression methods considering limitation of partial multicollinearity are relevant
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
