Theoretical properties of the ASMOD algorithm for empirical modelling (Q4353768)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1059638
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Theoretical properties of the ASMOD algorithm for empirical modelling |
scientific article; zbMATH DE number 1059638 |
Statements
Theoretical properties of the ASMOD algorithm for empirical modelling (English)
0 references
1 June 1998
0 references
model structure adaption
0 references
adaptive spline modelling of observation data
0 references
asymptotic convergence
0 references
B-spline
0 references
model complexity
0 references
consistency
0 references
0.85498357
0 references
0 references
0.83842945
0 references
0.8373759
0 references
0.8233044
0 references
0.8224923
0 references
Mathematical models generated from experimental data are considered. B-spline basis functions are used to construct a mapping between an input variable \(u\in \mathbb{R}^n\) and an output variable \(y\in\mathbb{R}\). The number of basis functions grows exponentially with the number of input variables. The ``adaptive spline modelling of observation data'' (ASMOD) algorithm decomposes the model into submodels. The model structure is adapted iteratively. The model complexity is reduced. In the paper theoretical aspects of the ASMOD algorithm are investigated. Bounds on the expected value of a criterion function used for model estimation are derived. Asymptotic convergence and consistency properties are analyzed.
0 references
