Between 1991 and 1995, the Environmental Monitoring and Assessment Program of the US Environmental Protection Agency conducted a survey of lakes in the Northeastern states of the US to determine the ecological condition of these waters. Here, to this end, we want to obtain estimates of the proportion of lakes at (high) risk of acidification or acidified already for each 8-digit hydrologic unit code (HUC) within the region of interest. Sample sizes for the 113 HUCs are very small and 27 HUCs are not even observed. Therefore, small area estimation techniques should be invoked for the estimation of the distribution function of acid neutralizing capacity (ANC) for each HUC. The procedure is based on a semiparametric M-quantile regression model in which ANC depends on elevation and the year of the survey linearly, and on the geographical position of the lake through an unknown smooth bivariate function estimated by low-rank thin plate splines. Copyright (C) 2008 John Wiley & Sons, Ltd.
Semiparametric M-quantile regression for estimating the proportion of acidic lakes in 8-digit HUCs of the Northeastern US
RANALLI, Maria Giovanna;
2008
Abstract
Between 1991 and 1995, the Environmental Monitoring and Assessment Program of the US Environmental Protection Agency conducted a survey of lakes in the Northeastern states of the US to determine the ecological condition of these waters. Here, to this end, we want to obtain estimates of the proportion of lakes at (high) risk of acidification or acidified already for each 8-digit hydrologic unit code (HUC) within the region of interest. Sample sizes for the 113 HUCs are very small and 27 HUCs are not even observed. Therefore, small area estimation techniques should be invoked for the estimation of the distribution function of acid neutralizing capacity (ANC) for each HUC. The procedure is based on a semiparametric M-quantile regression model in which ANC depends on elevation and the year of the survey linearly, and on the geographical position of the lake through an unknown smooth bivariate function estimated by low-rank thin plate splines. Copyright (C) 2008 John Wiley & Sons, Ltd.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
