post_parent): $temp_content = $post->post_content; $temp_content = explode("

",$temp_content); $temp_content = $temp_content[1]; $temp_content = explode("

",$temp_content); $temp_content = $temp_content[0]; $temp_content = strip_tags($temp_content); $temp_content = trim($temp_content); $authors = explode(",",$temp_content); ?> post_title));?>"> $value): ?> "> "> "> post_content); while ($parser->parse()) { if (($parser->iNodeName=="a")&&(substr_count($parser->iNodeAttributes['href'],".pdf")>0)): ?>

Univariate Non-Linear Regression Based on Spline Technology and Forsyth Normalized Orthogonal Polynomial

Pavlov A.A., Kalashnyk V.V., Khalymon A.Y.

The article describes method of univariate non-linear regression recovery on given range of argument value variation, with arbitrary distributed additional noise, upper estimate of which is considered to be known. Power polynomials are chosen as spline functions, which coefficients can be found using Forsyth normalized orthogonal polynomial based on theoretical and practical research, given in [2], or using special linear programming models. Also, substantiated constructive criterion, that ensures accurate solution of formulated problem, is given.


Full text (pdf)