Abstract
[Autom. eng. transl.] The purpose of this work is to point out the potential offered by Support Vector Machines (SVM) in the analysis of historical series. In the first section we introduce the SVM methodology starting from the generalization problem. In the second section, the generalization capabilities of SVM are tested through a simulation study and compared with those of neural networks. In the third section we apply SVM to a historical series that is well known in the literature in order to facilitate comparisons with alternative methods used in other works.
| Translated title of the contribution | [Autom. eng. transl.] Support Vector Machines and Historical Series Analysis |
|---|---|
| Original language | Italian |
| Title of host publication | MAF 2006 |
| Pages | 1-4 |
| Number of pages | 4 |
| Publication status | Published - 2006 |
| Event | CONVEGNO MAF 06 - Metodi Matematici e Statistici per le Assicurazioni e la Finanza - FISCIANO (SA) Duration: 11 Oct 2006 → 13 Oct 2006 |
Conference
| Conference | CONVEGNO MAF 06 - Metodi Matematici e Statistici per le Assicurazioni e la Finanza |
|---|---|
| City | FISCIANO (SA) |
| Period | 11/10/06 → 13/10/06 |
Keywords
- Nonparametric Regression
- Support Vector Machines
- Time Series
Fingerprint
Dive into the research topics of '[Autom. eng. transl.] Support Vector Machines and Historical Series Analysis'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver