Improving inference for nonlinear state-space models of animal\n population dynamics given biased sequential life stage data Article Swipe

MAKE IMPACT

Leo Polansky , Ken B. Newman , Lara Mitchell · YOU? · · 2019 · Open Access · · DOI: https://doi.org/10.48550/arxiv.1909.09111

State-space models (SSMs) are a popular tool for modeling animal abundances.\nInference difficulties for simple linear SSMs are well known, particularly in\nrelation to simultaneous estimation of process and observation variances.\nSeveral remedies to overcome estimation problems have been studied for\nrelatively simple SSMs, but whether these challenges and proposed remedies\napply for nonlinear stage-structured SSMs, an important class of ecological\nmodels, is less well understood. Here we identify improvements for inference\nabout nonlinear stage-structured SSMs fit with biased sequential life stage\ndata. Theoretical analyses indicate parameter identifiability requires\ncovariates in the state processes. Simulation studies show that plugging in\nexternally estimated observation variances, as opposed to jointly estimating\nthem with other parameters, reduces bias and standard error of estimates. In\ncontrast to previous results for simple linear SSMs, strong confounding between\njointly estimated process and observation variance parameters was not found in\nthe models explored here. However, when observation variance was also estimated\nin the motivating case study, the resulting process variance estimates were\nimplausibly low (near-zero). As SSMs are used in increasingly complex ways,\nunderstanding when inference can be expected to be successful, and what aids\nit, becomes more important. Our study illustrates (i) the need for relevant\nprocess covariates and (ii) the benefits of using externally estimated\nobservation variances for inference for nonlinear stage-structured SSMs.\n

Type

preprint

Landing Page

http://arxiv.org/abs/1909.09111

PDF

https://arxiv.org/pdf/1909.09111

OA Status

green

Related Works

10

OpenAlex ID

https://openalex.org/W4288104222

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4288104222

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.1909.09111

Digital Object Identifier

Title

Improving inference for nonlinear state-space models of animal\n population dynamics given biased sequential life stage data

Work title

Type

preprint

OpenAlex work type

Publication year

2019

Year of publication

Publication date

2019-09-19

Full publication date if available

Authors

Leo Polansky, Ken B. Newman, Lara Mitchell

List of authors in order

Landing page

https://arxiv.org/abs/1909.09111

Publisher landing page

PDF URL

https://arxiv.org/pdf/1909.09111

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/1909.09111

Direct OA link when available

Concepts

Identifiability, Covariate, Inference, Variance (accounting), Econometrics, Nonlinear system, Statistics, Population, Computer science, State space, Mathematics, Process (computing), Random effects model, Artificial intelligence, Operating system, Physics, Medicine, Quantum mechanics, Internal medicine, Meta-analysis, Accounting, Business, Sociology, Demography

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload