English

Semi-$G$-normal: a Hybrid between Normal and $G$-normal (Full Version)

Probability 2021-10-19 v3 Statistics Theory Statistics Theory

Abstract

The GG-expectation framework is a generalization of the classical probabilistic system motivated by Knightian uncertainty, where the GG-normal plays a central role. However, from a statistical perspective, GG-normal distributions look quite different from the classical normal ones. For instance, its uncertainty is characterized by a set of distributions which covers not only classical normal with different variances, but additional distributions typically having non-zero skewness. The GG-moments of GG-normals are defined by a class of fully nonlinear PDEs called GG-heat equations. To understand GG-normal in a probabilistic and stochastic way that is more friendly to statisticians and practitioners, we introduce a substructure called semi-GG-normal, which behaves like a hybrid between normal and GG-normal: it has variance uncertainty but zero-skewness. We will show that the non-zero skewness arises when we impose the GG-version sequential independence on the semi-GG-normal. More importantly, we provide a series of representations of random vectors with semi-GG-normal marginals under various types of independence. Each of these representations under a typical order of independence is closely related to a class of state-space volatility models with a common graphical structure. In short, semi-GG-normal gives a (conceptual) transition from classical normal to GG-normal, allowing us a better understanding of the distributional uncertainty of GG-normal and the sequential independence.

Keywords

Cite

@article{arxiv.2104.04910,
  title  = {Semi-$G$-normal: a Hybrid between Normal and $G$-normal (Full Version)},
  author = {Yifan Li and Reg Kulperger and Hao Yu},
  journal= {arXiv preprint arXiv:2104.04910},
  year   = {2021}
}

Comments

109 pages, 8 figures, a comprehensive document for conference and open discussions, to be divided later for publications, readers may navigate to the parts they are interested in by the table of contents

R2 v1 2026-06-24T01:02:45.947Z