Total least squares fitting Bass diffusion model

被引：11

作者：

Jukic, Dragan ^{[1
]}

机构：

[1] JJ Strossmayer Univ Osijek, Dept Math, HR-31000 Osijek, Croatia

来源：

MATHEMATICAL AND COMPUTER MODELLING | 2011年 / 53卷 / 9-10期

关键词：

Bass model; Total least squares; Total least squares estimate; Existence problem; Data fitting; PARAMETER-ESTIMATION PROBLEM; EXISTENCE; ALGORITHM; ERRORS;

D O I：

10.1016/j.mcm.2010.12.054

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

This paper is concerned with the Bass model, which is widely used as a first-purchase diffusion model in marketing research. Estimation of its parameters has been approached in the literature by various techniques. In this paper, we consider the nonlinear weighted total least squares (TLS) fitting approach. We show that it is possible that the TLS estimate does not exist. As a main result, two theorems on the existence of the total least squares estimate are obtained, as well as their generalization in the total l(s) norm (1 <= s < infinity). Several illustrative numerical examples are given to illustrate the efficiency of our approach. (C) 2010 Elsevier Ltd. All rights reserved.

引用

页码：1756 / 1770

页数：15

共 24 条

[1]

Atieg A., 2004, Australian and New Zealand Industrial and Applied Mathematics Journal, V45, pC187

[2]

Bailey N.T.J., 1957, The mathematical theory of epidemics

[3]

Bailey N. T. J., 1975, The Mathematical Theory of Infectious Diseases and Its Applications, V2nd

[4] NEW PRODUCT GROWTH FOR MODEL CONSUMER DURABLES [J].

BASS, FM .

MANAGEMENT SCIENCE SERIES A-THEORY, 1969, 15 (05) :215-227

[5] A STABLE AND EFFICIENT ALGORITHM FOR NONLINEAR ORTHOGONAL DISTANCE REGRESSION [J].

BOGGS, PT ;

BYRD, RH ;

SCHNABEL, RB .

SIAM JOURNAL ON SCIENTIFIC AND STATISTICAL COMPUTING, 1987, 8 (06) :1052-1078

[6] Criteria for unconstrained global optimization [J].

Demidenko, E. .

JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS, 2008, 136 (03) :375-395

[7]

Fuller W. A., 2009, Measurement error models

[8] AN ANALYSIS OF THE TOTAL LEAST-SQUARES PROBLEM [J].

GOLUB, GH ;

VANLOAN, CF .

SIAM JOURNAL ON NUMERICAL ANALYSIS, 1980, 17 (06) :883-893

[9] NUMERICAL ALGORITHMS FOR SOLVING NONLINEAR LP-NORM ESTIMATION PROBLEMS .1. A 1ST-ORDER GRADIENT ALGORITHM FOR WELL-CONDITIONED SMALL RESIDUAL PROBLEMS [J].

GONIN, R .

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 1986, 15 (03) :801-813

[10] Partial linearization of one class of the nonlinear total least squares problem by using the inverse model function [J].

Jukic, D ;

Scitovski, R ;

Späth, H .

COMPUTING, 1999, 62 (02) :163-178

← 1 2 3 →