Neural networks for parameter estimation in intractable models

The goal is to use deep learning models to estimate parameters in statistical models when standard likelihood estimation methods are computationally infeasible. For instance, inference for max-stable processes is exceptionally challenging even with small datasets, but simulation is straightforward....

Full description

Saved in:
Bibliographic Details
Published inComputational statistics & data analysis Vol. 185; no. C; p. 107762
Main Authors Lenzi, Amanda, Bessac, Julie, Rudi, Johann, Stein, Michael L.
Format Journal Article
LanguageEnglish
Published Netherlands Elsevier B.V 01.09.2023
Elsevier
Subjects
data analysis
data collection
Deep neural networks
Intractable likelihood
Max-stable distributions
Parameter estimation
statistics
Parameter estimation
Deep neural networks
Max-stable distributions
Intractable likelihood
Online AccessGet full text

Cover

More Information
Summary:The goal is to use deep learning models to estimate parameters in statistical models when standard likelihood estimation methods are computationally infeasible. For instance, inference for max-stable processes is exceptionally challenging even with small datasets, but simulation is straightforward. Data from model simulations are used to train deep neural networks and learn statistical parameters from max-stable models. The proposed neural network-based method provides a competitive alternative to current approaches, as demonstrated by considerable accuracy and computational time improvements. It serves as a proof of concept for deep learning in statistical parameter estimation and can be extended to other estimation problems.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
USDOE
ISSN:0167-9473
1872-7352
DOI:10.1016/j.csda.2023.107762