Proper account of long-term correlations in the observations improves state-space models' performances - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2024

Proper account of long-term correlations in the observations improves state-space models' performances

Nicolas Bez
Pierre Gloaguen
  • Fonction : Auteur
Marie-Pierre Etienne
  • Fonction : Auteur
Rocio Joo
  • Fonction : Auteur
Sophie Lanco
  • Fonction : Auteur
Etienne Rivot
  • Fonction : Auteur
Emily Walker
  • Fonction : Auteur
Mathieu Woillez
  • Fonction : Auteur
Stéphanie Mahévas
  • Fonction : Auteur

Résumé

State-space models are widely used in ecology to infer hidden behaviors. This study develops an extensive numerical simulation-estimation experiment to evaluate the state decoding accuracy of four simple state-space models. These models are obtained by combining different Markovian specifications (Markov and semi-Markov) for the hidden layer with the absence and presence of auto-correlation for the observation layer. Model parameters are issued from two sets of real annotated trajectories. Three metrics are developed to help interpret model performance. The first is the Hellinger distance between Markov and semi-Markov sojourn time probability distributions. The second is sensitive to the overlap between the probability density functions of state-dependent variables (e.g., speed variables). The third quantifies the deterioration of the inference conditions between AR0 and AR1 formulations. It emerges that the most sensitive model choice concerns the auto-correlation of the random processes describing the state-dependent variables. Opting for the absence of auto-correlation in the model while the state-dependent variables are actually auto-correlated, is detrimental to state decoding performance. Regarding the hidden layer, imposing a Markov structure while the state process is semi-Markov does not impair the state decoding performances. The real-life estimates are consistent with our experimental finding that performance deteriorates when there are significant temporal correlations that are not accounted for in the model. In light of these findings, we recommend that researchers carefully consider the structure of the statistical model they suggest and confirm its alignment with the process being modeled, especially when considering the auto-correlation of observed variables.
Fichier principal
Vignette du fichier
PCI_submitted_preprint_H_S_MM_afterProofReading.pdf (854.6 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Licence : CC BY - Paternité

Dates et versions

hal-04547315 , version 1 (15-04-2024)

Licence

Paternité

Identifiants

  • HAL Id : hal-04547315 , version 1

Citer

Nicolas Bez, Pierre Gloaguen, Marie-Pierre Etienne, Rocio Joo, Sophie Lanco, et al.. Proper account of long-term correlations in the observations improves state-space models' performances. 2024. ⟨hal-04547315⟩

Collections

IRD
0 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More