Spatial point process modelling and {Bayesian} inference for large data sets

@inproceedings{gillot_spatial_RM2024,
 abstract = {Modelling the galaxy distribution in our Universe is with no doubt a very important statistical challenge since the Universe contains around 200 billion galaxies. Among the typical available characteristics for the galaxies one must consider their position, mass, luminosity, and shape. Due to this, marked point processes appear as a natural modelling tool. There exists statistical methodology able to extract relevant information from marked point configurations. In this paper, we take the first step and propose to use non-parametric exploratory analysis and Bayesian posterior based inference in order to explore the first characteristic, namely the positions of more than 30000 galaxies. This is done in three steps. First, several windows of interest are selected. Then for each such window, a local exploratory analysis based on summary statistics is carried out. Finally, based on all the information gained in the previous steps, an appropriate model is fitted and posterior sampling is performed. Within this workflow, a new parametric multi-interaction point process model is introduced and fitted to the selected galaxy patterns. The quality of the estimation procedure and the significance of the estimated parameters is also assessed. Analysing several patterns allows us to have more insight into the stationary character of the entire observed data set and to depict perspectives with respect to the possible strategies for the general model fitting challenge.},
 author = {Gillot, N and Stoica, R S and Sarkka, A and Gemmerle, D},
 booktitle = {Proc. 2024 RING Meeting},
 language = {en},
 pages = {18},
 publisher = {ASGA},
 title = {Spatial point process modelling and {Bayesian} inference for large data sets},
 year = {2024}
}