Adaptive basis selection for functional data analysis via stochastic penalization

Autor(es)

Anselmo, Cezar A. F. ; Dias, Ronaldo ; Garcia, Nancy L.

• Tipo de documento: Artigo Científico
• Fonte: COMPUTATIONAL & APPLIED MATHEMATICS; v. 24, n. 2, p. 21-pg., 2005-01-01.

Resumo

We propose an adaptive method of analyzing a collection of curves which can be, individually, modeled as a linear combination of spline basis functions. Through the introduction of latent Bernoulli variables, the number of basis functions, the variance of the error measurements and the coefficients of the expansion are determined. We provide a modification of the stochastic EM algorithm for which numerical results show that the estimates are very close to the true curve in the sense of L-2 norm. (AU)

• Processo FAPESP: 01/00258-0 - Selecao adaptativa de funcoes de base em analise funcional de dados via penalizacao estocastica.
• Beneficiário: Cezar Augusto de Freitas Anselmo
• Modalidade de apoio: Bolsas no Brasil - Mestrado
• Processo FAPESP: 02/01554-5 - Procedimentos adaptativos em análise funcional de dados
• Beneficiário: Ronaldo Dias
• Modalidade de apoio: Auxílio à Pesquisa - Regular