Activity Number:

436
 SPEED: Tests, Trials, Biomarkers, and Other Topics in Biometrics

Type:

Contributed

Date/Time:

Tuesday, July 31, 2018 : 3:05 PM to 3:50 PM

Sponsor:

Biometrics Section

Abstract #332870


Title:

Adjusting a Finite Population Block Kriging Estimator for Imperfect Detection

Author(s):

Matthew Higham*

Companies:


Keywords:

geostatistics;
sampling;
spatial

Abstract:

A finite population version of block kriging (FPBK) that assumes perfect detection is routinely used to estimate moose abundance in Alaska. We consider two extensions of FPBK to incorporate imperfect moose detection for estimates of population total and their standard errors. The Spatial Population Estimator with Detection: Ratio then Add (SPEDRA) estimator adjusts the observed moose counts for each sample unit by the sample unit's estimated detection probability prior to spatial modeling. The Spatial Population Estimator with Detection: Add then Ratio (SPEDAR) estimator uses spatial modeling first on the observed moose counts; the estimator is then adjusted by the estimated mean detection probability. We find that both estimators perform similarly in simulations with respect to both bias and mean square prediction error and give comparable estimates of the moose population total when applied to data from moose surveys in Togiak National Wildlife Refuge (AK). The issue in whether or not to divide by a detection probability sitewise or to divide by the mean detection probability across all sampled sites appears in many population abundance estimation problems.

Authors who are presenting talks have a * after their name.
Back to the full JSM 2018 program
