Dice Question Streamline Icon: https://streamlinehq.com

Complex-manifold (Kähler) structures for statistical models

Investigate the Fisher–Rao geometry of parametric statistical models M={p_θ: θ∈Θ} as complex manifolds when the model order is even, including the existence and properties of associated Kähler structures.

Information Square Streamline Icon: https://streamlinehq.com

Background

The work suggests moving beyond real manifold structures to complex manifolds for even-dimensional models, motivated by connections to Kähler geometry and prior preliminary studies. Establishing a coherent complex (and potentially Kähler) geometric framework for Fisher–Rao manifolds could open new analytical tools and structural insights, complementing existing real and Hessian geometries.

References

Finally, let us list some open problems: Problem 4. Study $M={p_\theta:\theta\in\Theta}$ as a complex manifold instead of a real manifold when the statistical model $m$ is of even order. See preliminary work in and K"ahler geometry in information geometry.

Approximation and bounding techniques for the Fisher-Rao distances between parametric statistical models (2403.10089 - Nielsen, 15 Mar 2024) in Conclusion: Summary and some open problems (final section)