English

Shrinkage priors on complex-valued circular-symmetric autoregressive processes

Statistics Theory 2021-02-05 v3 Statistics Theory

Abstract

We investigate shrinkage priors on power spectral densities for complex-valued circular-symmetric autoregressive processes. We construct shrinkage predictive power spectral densities, which asymptotically dominate (i) the Bayesian predictive power spectral density based on the Jeffreys prior and (ii) the estimative power spectral density with the maximal likelihood estimator, where the Kullback-Leibler divergence from the true power spectral density to a predictive power spectral density is adopted as a risk. Furthermore, we propose general constructions of objective priors for K\"ahler parameter spaces, utilizing a positive continuous eigenfunction of the Laplace-Beltrami operator with a negative eigenvalue. We present numerical experiments on a complex-valued stationary autoregressive model of order 11.

Cite

@article{arxiv.2004.02389,
  title  = {Shrinkage priors on complex-valued circular-symmetric autoregressive processes},
  author = {Hidemasa Oda and Fumiyasu Komaki},
  journal= {arXiv preprint arXiv:2004.02389},
  year   = {2021}
}

Comments

revised; Figures are modified

R2 v1 2026-06-23T14:40:22.335Z