Showing 1–4 of 4 results for author: Christopoulos, D T

On the effective radius of convergence for a given truncated power series expansion

Authors: Demetris T. Christopoulos

Abstract: An effective radius of convergence is defined and computed for any truncated Taylor series. Applications to well known series are performed and is shown that a range of good coincidence for actual and approximative plot can always be found. For sufficient large degree of approximation the effective radius is also an estimation of the true non-infinite radius of convergence. An effective radius of convergence is defined and computed for any truncated Taylor series. Applications to well known series are performed and is shown that a range of good coincidence for actual and approximative plot can always be found. For sufficient large degree of approximation the effective radius is also an estimation of the true non-infinite radius of convergence. △ Less

Submitted 9 December, 2013; originally announced December 2013.

Comments: 11 pages, 7 figures

MSC Class: Primary 41A58; Secondary 40A05

Polynomial regression using trapezoidal rule for computing Legendre coefficients

Authors: Demetris T. Christopoulos

Abstract: We are presenting a method for computing the Fourier coefficients of a given polynomial regression by using the trapezoidal rule for numerical integration. As function basis we use the orthogonal Legendre polynomials. The results are accurate and stable compared to Forsythe's method. We are presenting a method for computing the Fourier coefficients of a given polynomial regression by using the trapezoidal rule for numerical integration. As function basis we use the orthogonal Legendre polynomials. The results are accurate and stable compared to Forsythe's method. △ Less

Submitted 29 November, 2013; originally announced November 2013.

Comments: 13 pages, 2 figures, 4 tables

MSC Class: Primary 62J05; Secondary 65D99

Linear Regression without computing pseudo-inverse matrix

Authors: Demetris T. Christopoulos

Abstract: We are presenting a method of linear regression based on Gram-Schmidt orthogonal projection that does not compute a pseudo-inverse matrix. This is useful when we want to make several regressions with random data vectors for simulation purposes. We are presenting a method of linear regression based on Gram-Schmidt orthogonal projection that does not compute a pseudo-inverse matrix. This is useful when we want to make several regressions with random data vectors for simulation purposes. △ Less

Submitted 6 November, 2013; originally announced November 2013.

Comments: 4 pages

MSC Class: Primary 97K80; Secondary 62J05

Developing methods for identifying the inflection point of a convex/concave curve

Authors: Demetris T. Christopoulos

Abstract: We are introducing two methods for revealing the true inflection point of data that contains or not error. The starting point is a set of geometrical properties that follow the existence of an inflection point p for a smooth function. These properties connect the concept of convexity/concavity before and after p respectively with three chords defined properly. Finally a set of experiments is prese… ▽ More We are introducing two methods for revealing the true inflection point of data that contains or not error. The starting point is a set of geometrical properties that follow the existence of an inflection point p for a smooth function. These properties connect the concept of convexity/concavity before and after p respectively with three chords defined properly. Finally a set of experiments is presented for the class of sigmoid curves and for the third order polynomials. △ Less

Submitted 4 August, 2014; v1 submitted 24 June, 2012; originally announced June 2012.

Comments: 10 pages, 6 figures; 29 pages, 10 figures, 20 tables, new layout, new definitions in section 1, graphical explanation of ESE & EDE methods, improved bibliography

MSC Class: Primary 65H99; Secondary 62F12