Towards optimal Takacs–Fiksel estimation

Jean François Coeurjolly; Yongtao Guan; Mahdieh Khanmohammadi; Rasmus Waagepetersen

doi:10.1016/j.spasta.2016.08.002

Towards optimal Takacs–Fiksel estimation

Jean François Coeurjolly^*, Yongtao Guan, Mahdieh Khanmohammadi, Rasmus Waagepetersen

^*Corresponding author for this work

Department of Computer Science

2 Citations (Scopus)

Abstract

The Takacs–Fiksel method is a general approach to estimate the parameters of a spatial Gibbs point process. This method embraces standard procedures such as the pseudolikelihood and is defined via weight functions. In this paper we propose a general procedure to find weight functions which reduce the Godambe information and thus outperform pseudolikelihood in certain situations. The new procedure is applied to a standard dataset and to a recent neuroscience replicated point pattern dataset. Finally, the performance of the new procedure is investigated in a simulation study.

Original language	English
Journal	Spatial Statistics
Volume	18
Issue number	Part B
Pages (from-to)	396-411
Number of pages	16
ISSN	2211-6753
DOIs	https://doi.org/10.1016/j.spasta.2016.08.002
Publication status	Published - 2016

Keywords

Gibbs point processes
Godambe information
Optimal estimation
Pseudolikelihood
Spatial point processes

Access to Document

10.1016/j.spasta.2016.08.002

Cite this

@article{f0ed65bce373495f919a9a1a7de762c2,

title = "Towards optimal Takacs–Fiksel estimation",

abstract = "The Takacs–Fiksel method is a general approach to estimate the parameters of a spatial Gibbs point process. This method embraces standard procedures such as the pseudolikelihood and is defined via weight functions. In this paper we propose a general procedure to find weight functions which reduce the Godambe information and thus outperform pseudolikelihood in certain situations. The new procedure is applied to a standard dataset and to a recent neuroscience replicated point pattern dataset. Finally, the performance of the new procedure is investigated in a simulation study.",

keywords = "Gibbs point processes, Godambe information, Optimal estimation, Pseudolikelihood, Spatial point processes",

author = "Coeurjolly, {Jean Fran{\c c}ois} and Yongtao Guan and Mahdieh Khanmohammadi and Rasmus Waagepetersen",

year = "2016",

doi = "10.1016/j.spasta.2016.08.002",

language = "English",

volume = "18",

pages = "396--411",

journal = "Spatial Statistics",

issn = "2211-6753",

publisher = "Elsevier",

number = "Part B",

}

TY - JOUR

T1 - Towards optimal Takacs–Fiksel estimation

AU - Coeurjolly, Jean François

AU - Guan, Yongtao

AU - Khanmohammadi, Mahdieh

AU - Waagepetersen, Rasmus

PY - 2016

Y1 - 2016

N2 - The Takacs–Fiksel method is a general approach to estimate the parameters of a spatial Gibbs point process. This method embraces standard procedures such as the pseudolikelihood and is defined via weight functions. In this paper we propose a general procedure to find weight functions which reduce the Godambe information and thus outperform pseudolikelihood in certain situations. The new procedure is applied to a standard dataset and to a recent neuroscience replicated point pattern dataset. Finally, the performance of the new procedure is investigated in a simulation study.

AB - The Takacs–Fiksel method is a general approach to estimate the parameters of a spatial Gibbs point process. This method embraces standard procedures such as the pseudolikelihood and is defined via weight functions. In this paper we propose a general procedure to find weight functions which reduce the Godambe information and thus outperform pseudolikelihood in certain situations. The new procedure is applied to a standard dataset and to a recent neuroscience replicated point pattern dataset. Finally, the performance of the new procedure is investigated in a simulation study.

KW - Gibbs point processes

KW - Godambe information

KW - Optimal estimation

KW - Pseudolikelihood

KW - Spatial point processes

U2 - 10.1016/j.spasta.2016.08.002

DO - 10.1016/j.spasta.2016.08.002

M3 - Journal article

AN - SCOPUS:84995743991

SN - 2211-6753

VL - 18

SP - 396

EP - 411

JO - Spatial Statistics

JF - Spatial Statistics

IS - Part B

ER -

Towards optimal Takacs–Fiksel estimation

Abstract

Keywords

Access to Document

Fingerprint

Cite this