The Selective Regularization of a Linear Regression Model: A Recent Study| Chapter 10 | Novel Research Aspects in Mathematical and Computer Science Vol. 7

 This article discusses how to build a linear regression model that includes regularising the system matrix of normal equations. Unlike the traditional ridge regression, which adds positive parameters to all of a matrix's diagonal terms, only the matrix diagonal entries that correspond to the data with a strong correlation are enhanced. As a result, the matrix conditioning decreases, which also affects the coefficients of the associated regression equation. The choice of the entries to be increased is based on the triangular decomposition of the correlation matrix of the original dataset. The effectiveness of the strategy is assessed on a known dataset using both ridge regression and the results of using the well-known algorithms LARS and Lasso.

Author(s) Details:

V. N. Lutay,
Institute of Computer Technologies and Information Security, Southern Federal University, Chekhov Street, 2, Taganrog - 347922, Russia.

N. S. Khusainov,
Institute of Computer Technologies and Information Security, Southern Federal University, Chekhov Street, 2, Taganrog - 347922, Russia.

Please see the link here: https://stm.bookpi.org/NRAMCS-V7/article/view/7955