Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Artículo científico |
| Lingua: | en |
| Pubblicazione: |
Universidade Estadual de Maringá
2015
|
| Soggetti: | |
| Accesso online: | https://www.redalyc.org/articulo.oa?id=303241625014 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
Sommario:
- Is the Conditional Density Netw ork more suitable than the Maximum likelihood for fitting the Generalized Extreme Value Distribution? Monica Cristina Meschiatti Gabriel Constantino Blain Ingeniería sample size Neural network extreme precipitation The Generalized Extreme value Distribution (G EV) has been widely used to assess the probability of extreme weather events and the paramete r estimation method is a key factor for improving its quantile estimates. On such background, this st udy aimed to indicate under which conditions (sample size and tail behavior) the Conditional Density Network (CDN) leads to better GEV quantile estimates than the widely used Maximum likelihood method (MLE) does. With Monte Carlo simulations and rainfall series of several Brazilians regions, we highlig ht the following results: th e return period and the tail behavior of the GEV (specified by the shape parameter) are two of the main factors affecting the quantile estimates. For -0.1 ≤ shape ≤ 0.1 and sample size ≤ 50, the CDN outperformed the MLE. For shape ≥ 0.20 the CDN outperformed the MLE for all sample sizes (30-90). The results also suggested that the CDN is more suitable than the MLE for fitting the GEV parameter to the Brazilian extreme rainfall series. We conclude that when the shape parameter ar e equal to or greater than -0.1 the CDN should be preferred over the MLE. 2015 artículo científico 1806-2563 https://www.redalyc.org/articulo.oa?id=303241625014 en http://www.redalyc.org/revista.oa?id=3032 Acta Scientiarum. Technology application/pdf Universidade Estadual de Maringá Acta Scientiarum. Technology (Brasil) Num.4 Vol.37