Diffusion means in geometric spaces
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Diffusion means in geometric spaces. / Eltzner, Benjamin; Hansen, Pernille E.H.; Huckemann, Stephan F.; Sommer, Stefan.
In: Bernoulli, Vol. 29, No. 4, 2023, p. 3141-3170.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Diffusion means in geometric spaces
AU - Eltzner, Benjamin
AU - Hansen, Pernille E.H.
AU - Huckemann, Stephan F.
AU - Sommer, Stefan
N1 - Publisher Copyright: © 2023 ISI/BS.
PY - 2023
Y1 - 2023
N2 - We introduce a location statistic for distributions on non-linear geometric spaces, the diffusion mean, serving as an extension and an alternative to the Fréchet mean. The diffusion mean arises as the generalization of Gaussian maximum likelihood analysis to non-linear spaces by maximizing the likelihood of a Brownian motion. The diffusion mean depends on a time parameter t, which admits the interpretation of the allowed variance of the diffusion. The diffusion t-mean of a distribution X is the most likely origin of a Brownian motion at time t, given the end-point distribution X. We give a detailed description of the asymptotic behavior of the diffusion estimator and provide sufficient conditions for the diffusion estimator to be strongly consistent. Particularly, we present a smeary central limit theorem for diffusion means and we show that joint estimation of the mean and diffusion variance rules out smeariness in all directions simultaneously in general situations. Furthermore, we investigate properties of the diffusion mean for distributions on the sphere Sm. Experimentally, we consider simulated data and data from magnetic pole reversals, all indicating similar or improved convergence rate compared to the Fréchet mean. Here, we additionally estimate t and consider its effects on smeariness and uniqueness of the diffusion mean for distributions on the sphere.
AB - We introduce a location statistic for distributions on non-linear geometric spaces, the diffusion mean, serving as an extension and an alternative to the Fréchet mean. The diffusion mean arises as the generalization of Gaussian maximum likelihood analysis to non-linear spaces by maximizing the likelihood of a Brownian motion. The diffusion mean depends on a time parameter t, which admits the interpretation of the allowed variance of the diffusion. The diffusion t-mean of a distribution X is the most likely origin of a Brownian motion at time t, given the end-point distribution X. We give a detailed description of the asymptotic behavior of the diffusion estimator and provide sufficient conditions for the diffusion estimator to be strongly consistent. Particularly, we present a smeary central limit theorem for diffusion means and we show that joint estimation of the mean and diffusion variance rules out smeariness in all directions simultaneously in general situations. Furthermore, we investigate properties of the diffusion mean for distributions on the sphere Sm. Experimentally, we consider simulated data and data from magnetic pole reversals, all indicating similar or improved convergence rate compared to the Fréchet mean. Here, we additionally estimate t and consider its effects on smeariness and uniqueness of the diffusion mean for distributions on the sphere.
KW - Diffusion mean
KW - generalized Fréchet mean
KW - geometric statistics
KW - maximum likelihood estimation
KW - spherical statistics
UR - http://www.scopus.com/inward/record.url?scp=85161632699&partnerID=8YFLogxK
U2 - 10.3150/22-BEJ1578
DO - 10.3150/22-BEJ1578
M3 - Journal article
AN - SCOPUS:85161632699
VL - 29
SP - 3141
EP - 3170
JO - Bernoulli
JF - Bernoulli
SN - 1350-7265
IS - 4
ER -
ID: 366982377