Parameter Identifiability and Redundancy: Theoretical Considerations

Mark P. Little, Wolfgang F. Heidenreich, Guangquan Li · 2010

Read Parameter Identifiability and Redundancy: Theoretical Considerations by Mark P. Little, Wolfgang F. Heidenreich, Guangquan Li (2010) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0008915.txt.

Description

Mark Little, Wolfgang Heidenreich, and Guangquan Li laid out theory for when model parameters can actually be known. Their paper appeared in PLOS ONE in 2010. Models of complex biology can carry dozens of parameters. Some cannot be pinned down from data, no matter how much is gathered. Such parameters are called unidentifiable. Redundant parameters can also be merged without changing predictions. Modelers often discover these traps only after wasted effort.

Their theoretical treatment gives tests to spot trouble early. Checking identifiability before fitting saves time and false confidence. Any field leaning on big fitted models benefits, from epidemiology to cell biology. The paper serves as a rulebook for honest modeling.

Catalog record

CREATORMark P. Little, Wolfgang F. Heidenreich, Guangquan Li
YEAR2010
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27 KB
LICENSECreative Commons Attribution License
RIGHTSCC BY
SOURCEPLOS ONE

Source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0008915

Browse the full library · About the Tower of Babel