Statistical Issues in Fisheries' Stock Assessments

Decisions concerning the management of fisheries are founded on confidence statements for interest parameters such as biomass and exploitation rate, derived from complex structural models that describe the dynamics of fisheries. We identify four generic statistical issues and focus on how they impac...

Full description

Saved in:
Bibliographic Details
Published in:Scandinavian journal of statistics 2002-06, Vol.29 (2), p.245-267
Main Authors: GAVARIS, STRATIS, IANELLI, JAMES N.
Format: Article
Language:English
Subjects:
bootstrap
confidence distribution
Fisheries
Fisheries management
Fisheries science
fisheries' assessment
history of fisheries' problem
inference
MCMC
model selection
Modeling
Mortality
Ocean fisheries
Oceans
parameter indeterminacy
Parametric models
Publishing industry
Term weighting
weighting
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Decisions concerning the management of fisheries are founded on confidence statements for interest parameters such as biomass and exploitation rate, derived from complex structural models that describe the dynamics of fisheries. We identify four generic statistical issues and focus on how they impact on the reliability of those confidence statements: (a) parameters for which the data have little or no information; (b) competing structural relationships; (c) weighting of observations; and (d) alternative methods for computing confidence statements. Our purpose is to give an exposition of how these issues impact on fisheries' analyses, with the intent of stimulating thought on more effective alternatives. We describe the fisheries' management context and use two specific studies to illustrate how these generic statistical issues impact on fisheries assessment results. It is demonstrated that these statistical issues can have a profound impact on fishery management decisions and that established approaches to handle them have not been fully developed.
ISSN:0303-6898
1467-9469
DOI:10.1111/1467-9469.00282