Premium
A non‐linear and non‐Gaussian state‐space model for censored air pollution data
Author(s) -
Johns Craig J.,
Shumway Robert H.
Publication year - 2005
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.693
Subject(s) - lidar , markov chain monte carlo , censoring (clinical trials) , gaussian , gaussian network model , environmental science , particle filter , monte carlo method , computer science , statistics , remote sensing , meteorology , mathematics , geography , kalman filter , physics , quantum mechanics
Lidar technology is used to quantify airborne particulate matter less than 10 μm in diameter (PM 10 ). These spatio‐temporal lidar data on PM 10 are subject to censoring due to detection limits. A non‐linear and non‐Gaussian state‐space model is modified to accommodate data subject to detection limits and outline strategies for Markov‐chain Monte Carlo estimation and filtering. The methods are applied to spatio‐temporal lidar measurements of dust particle concentrations. Copyright © 2004 John Wiley & Sons, Ltd.