This paper introduces a novel Hermite spline model for data regression, integrating both function values and derivatives along with a penalty term to control smoothness. A comparative analysis is conducted with conventional penalized models, specifically with P-spline models. The primary objective of this study is to empirically demonstrate the superior performance of the proposed model in reconstructing data, even in the absence of a penalty term (pure regression).
A Hermite spline model for data regression / Campagna, R.; Cotronei, M.; Fazzino, D.. - In: MATHEMATICS AND COMPUTERS IN SIMULATION. - ISSN 0378-4754. - 229:(2025), pp. 222-234. [10.1016/j.matcom.2024.09.011]
A Hermite spline model for data regression
Cotronei M.
;Fazzino D.
2025-01-01
Abstract
This paper introduces a novel Hermite spline model for data regression, integrating both function values and derivatives along with a penalty term to control smoothness. A comparative analysis is conducted with conventional penalized models, specifically with P-spline models. The primary objective of this study is to empirically demonstrate the superior performance of the proposed model in reconstructing data, even in the absence of a penalty term (pure regression).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
