In this paper, kernel-type estimators of the spatial distribution function are constructed, under non-constant trend, by first approximating the distribution at the sampled sites and then obtaining a weighted average of the resulting values. Unlike other alternatives, our proposals provide non- decreasing functions and do not require previous estimations of the indicator variogram or the trend function. However, appropriate bandwidths parameters are needed and selection of them in practice will be addressed.
(2014). Construction of probability maps under local stationarity [conference presentation - intervento a convegno]. Retrieved from http://hdl.handle.net/10446/31692
Construction of probability maps under local stationarity
2014-01-01
Abstract
In this paper, kernel-type estimators of the spatial distribution function are constructed, under non-constant trend, by first approximating the distribution at the sampled sites and then obtaining a weighted average of the resulting values. Unlike other alternatives, our proposals provide non- decreasing functions and do not require previous estimations of the indicator variogram or the trend function. However, appropriate bandwidths parameters are needed and selection of them in practice will be addressed.| File | Dimensione del file | Formato | |
|---|---|---|---|
|
3132-6585-1-PB.pdf
accesso aperto
Descrizione: publisher's version - versione dell'editore
Dimensione del file
135.17 kB
Formato
Adobe PDF
|
135.17 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo
