ML Anthology

On the Asymptotic Distribution of the Least-Squares Estimators in Unidentifiable Models

Taichi Hayasaka, Masashi Kitahara, Shiro Usui
NeCo 2004 pp. 99-114
doi:10.1162/08997660460734010 /neco/2004/hayasaka2004neco-asymptotic/

Abstract

In order to analyze the stochastic property of multilayered perceptrons or other learning machines, we deal with simpler models and derive the asymptotic distribution of the least-squares estimators of their parameters. In the case where a model is unidentified, we show different results from traditional linear models: the well-known property of asymptotic normality never holds for the estimates of redundant parameters.

NeCo Semantic Scholar

Cite

Text

Hayasaka et al. "On the Asymptotic Distribution of the Least-Squares Estimators in Unidentifiable Models." Neural Computation, 2004. doi:10.1162/08997660460734010

Markdown

[Hayasaka et al. "On the Asymptotic Distribution of the Least-Squares Estimators in Unidentifiable Models." Neural Computation, 2004.](https://mlanthology.org/neco/2004/hayasaka2004neco-asymptotic/) doi:10.1162/08997660460734010

BibTeX

@article{hayasaka2004neco-asymptotic,
  title     = {{On the Asymptotic Distribution of the Least-Squares Estimators in Unidentifiable Models}},
  author    = {Hayasaka, Taichi and Kitahara, Masashi and Usui, Shiro},
  journal   = {Neural Computation},
  year      = {2004},
  pages     = {99-114},
  doi       = {10.1162/08997660460734010},
  volume    = {16},
  url       = {https://mlanthology.org/neco/2004/hayasaka2004neco-asymptotic/}
}