Resumen

POLANCO, Maryorys Verónica. Curve smoothing for fitting data with nonlinear structure from a p-splines perspective. Ingenium et Potentia [online]. 2024, vol.6, n.11, pp.69-94.  Epub 31-Oct-2024. ISSN 2665-0304.  https://doi.org/10.35381/i.p.v6i11.4158.

The objective of this work was to conduct a theoretical, descriptive, and applied documentary approach to the P-splines approximation theory. The methodology employed included, firstly, a literature review to identify the main characteristics of the method and existing approaches; secondly, an experimental stage consisting of fitting six data series from a study on the population dynamics of Ralstonia Solanacearum in tomato plants using the R-Project software; and finally, the obtained results. It was evident that the parameter λ is responsible for controlling the smoothing, and in the search for a value for this parameter that optimizes the fit, one can rely on the estimate provided by the GAM function, as it provided almost identical fits to the models with an assigned value of λ, which, according to the considered criteria and indices, were selected as the final model.

Palabras clave : Statistical analysis; smoothing; data; series.

        · resumen en Español     · texto en Español     · Español ( pdf )