Semiparametric Maximum Likelihood for Nonlinear Regression with Measurement Errors

This article demonstrates semiparametric maximum likelihood estimation of a nonlinear growth model for fish lengths using imprecisely measured ages. Data on the species corvina reina, found in the Gulf of Nicoya, Costa Rica, consist of lengths and imprecise ages for 168 fish and precise ages for a s...

Full description

Saved in:
Bibliographic Details
Published in:Biometrics 2002-06, Vol.58 (2), p.448-453
Main Authors: Suh, Eun‐Young, Schafer, Daniel W
Format: Article
Language:English
Subjects:
Algorithms
Animals
Biometrics
Biometry
Consultant's Forum
Data Interpretation, Statistical
EM algorithm
Errors-in-variables
Estimation methods
Estimators
fish
Fisheries - statistics & numerical data
Fishes - growth & development
growth models
Growth-curve analysis
Internal validation
Likelihood Functions
Maximum likelihood estimation
Maximum likelihood estimators
Measurement error model regression
Mixture model
Modeling
Nonlinear Dynamics
Nonparametric maximum likelihood
Nonparametric models
Regression Analysis
Sciaenidae
Structural model
Vertex direction methods
Vertices
von Bertalanffy model
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:This article demonstrates semiparametric maximum likelihood estimation of a nonlinear growth model for fish lengths using imprecisely measured ages. Data on the species corvina reina, found in the Gulf of Nicoya, Costa Rica, consist of lengths and imprecise ages for 168 fish and precise ages for a subset of 16 fish. The statistical problem may therefore be classified as nonlinear errors‐in‐variables regression with internal validation data. Inferential techniques are based on ideas extracted from several previous works on semiparametric maximum likelihood for errors‐in‐variables problems. The illustration of the example clarifies practical aspects of the associated computational, inferential, and data analytic techniques.
ISSN:0006-341X
1541-0420
DOI:10.1111/j.0006-341X.2002.00448.x