Premium
Multivariate density estimation for interval‐censored data with application to a forest fire modelling problem
Author(s) -
Braun W. J.,
Stafford J. E.
Publication year - 2016
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/env.2396
Subject(s) - multivariate statistics , estimator , statistics , density estimation , interval estimation , mathematics , interval (graph theory) , estimation , context (archaeology) , econometrics , confidence interval , geography , engineering , systems engineering , archaeology , combinatorics
Local likelihood‐based density estimation methods are developed for multivariate interval‐censored data. An extension of the Nadaraya–Watson local regression estimator to the interval‐censored data context arises in a natural way. The conditional density estimation scheme is used to study the holdover distribution of lightning‐caused wildfire ignitions in Northern Ontario. Copyright © 2016 John Wiley & Sons, Ltd.