The maximum likelihood degree of toric varieties
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
The maximum likelihood degree of toric varieties. / Améndola, Carlos; Bliss, Nathan; Burke, Isaac; Gibbons, Courtney R.; Helmer, Martin; Hoşten, Serkan; Nash, Evan D.; Rodriguez, Jose Israel; Smolkin, Daniel.
In: Journal of Symbolic Computation, Vol. 92, 2018, p. 222-242.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - The maximum likelihood degree of toric varieties
AU - Améndola, Carlos
AU - Bliss, Nathan
AU - Burke, Isaac
AU - Gibbons, Courtney R.
AU - Helmer, Martin
AU - Hoşten, Serkan
AU - Nash, Evan D.
AU - Rodriguez, Jose Israel
AU - Smolkin, Daniel
PY - 2018
Y1 - 2018
N2 - We study the maximum likelihood (ML) degree of toric varieties, known as discrete exponential models in statistics. By introducing scaling coefficients to the monomial parameterization of the toric variety, one can change the ML degree. We show that the ML degree is equal to the degree of the toric variety for generic scalings, while it drops if and only if the scaling vector is in the locus of the principal A-determinant. We also illustrate how to compute the ML estimate of a toric variety numerically via homotopy continuation from a scaled toric variety with low ML degree. Throughout, we include examples motivated by algebraic geometry and statistics. We compute the ML degree of rational normal scrolls and a large class of Veronese-type varieties. In addition, we investigate the ML degree of scaled Segre varieties, hierarchical log-linear models, and graphical models.
AB - We study the maximum likelihood (ML) degree of toric varieties, known as discrete exponential models in statistics. By introducing scaling coefficients to the monomial parameterization of the toric variety, one can change the ML degree. We show that the ML degree is equal to the degree of the toric variety for generic scalings, while it drops if and only if the scaling vector is in the locus of the principal A-determinant. We also illustrate how to compute the ML estimate of a toric variety numerically via homotopy continuation from a scaled toric variety with low ML degree. Throughout, we include examples motivated by algebraic geometry and statistics. We compute the ML degree of rational normal scrolls and a large class of Veronese-type varieties. In addition, we investigate the ML degree of scaled Segre varieties, hierarchical log-linear models, and graphical models.
KW - A-discriminant
KW - Maximum likelihood degree
KW - Toric variety
UR - http://www.scopus.com/inward/record.url?scp=85045325960&partnerID=8YFLogxK
U2 - 10.1016/j.jsc.2018.04.016
DO - 10.1016/j.jsc.2018.04.016
M3 - Journal article
AN - SCOPUS:85045325960
VL - 92
SP - 222
EP - 242
JO - Journal of Symbolic Computation
JF - Journal of Symbolic Computation
SN - 0747-7171
ER -
ID: 199804620