Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Recurso digital |
| Lenguaje: | |
| Publicado: |
Zenodo
2026
|
| Acceso en línea: | https://doi.org/10.5281/zenodo.18412250 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
Tabla de Contenidos:
- <p>Readme</p> <p>The code computes archetypes of a data matrix X and provides r archetypes. The default value is r=4.</p> <p>The code requires essntially the data matrix X. X could be 2-dimensional, say nm, where n is the sample size, and m is the number of variables. It could also be 3-dimensional for example, nla*lo, for a field, in which n again is the sample size, and la is the latitudinal dimension and lo is the longitudinal dimension</p> <p>Example: [A B Za Zb] = aa_dynsys (randn(100,2));</p> <p>A and B are weight matrices Zb are the archetypes, Za are similar to archetypes but have smaller amplitudes (referred to in the ms as duals).</p> <p>In some cases the matrix may be scaled to get smoother convergence.</p>