Leave-group-out cross-validation for latent Gaussian models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2022
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911035460419584 |
|---|---|
| author | Liu, Zhedong Van Niekerk, Janet Rue, Haavard |
| author_facet | Liu, Zhedong Van Niekerk, Janet Rue, Haavard |
| contents | Evaluating the predictive performance of a statistical model is commonly done using cross-validation. Among the various methods, leave-one-out cross-validation (LOOCV) is frequently used. Originally designed for exchangeable observations, LOOCV has since been extended to other cases such as hierarchical models. However, it focuses primarily on short-range prediction and may not fully capture long-range prediction scenarios. For structured hierarchical models, particularly those involving multiple random effects, the concepts of short- and long-range predictions become less clear, which can complicate the interpretation of LOOCV results. In this paper, we propose a complementary cross-validation framework specifcally tailored for longer-range prediction in latent Gaussian models, including those with structured random effects. Our approach differs from LOOCV by excluding a carefully constructed set from the training set, which better emulates longer-range prediction conditions. Furthermore, we achieve computational effciency by adjusting the full joint posterior for this modifed cross-validation, thus eliminating the need for model reftting. This method is implemented in the R-INLA package (www.r-inla.org) and can be adapted to a variety of inferential frameworks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2210_04482 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Leave-group-out cross-validation for latent Gaussian models Liu, Zhedong Van Niekerk, Janet Rue, Haavard Computation Applications Methodology 62-04 62C10 62F15 62J12 Evaluating the predictive performance of a statistical model is commonly done using cross-validation. Among the various methods, leave-one-out cross-validation (LOOCV) is frequently used. Originally designed for exchangeable observations, LOOCV has since been extended to other cases such as hierarchical models. However, it focuses primarily on short-range prediction and may not fully capture long-range prediction scenarios. For structured hierarchical models, particularly those involving multiple random effects, the concepts of short- and long-range predictions become less clear, which can complicate the interpretation of LOOCV results. In this paper, we propose a complementary cross-validation framework specifcally tailored for longer-range prediction in latent Gaussian models, including those with structured random effects. Our approach differs from LOOCV by excluding a carefully constructed set from the training set, which better emulates longer-range prediction conditions. Furthermore, we achieve computational effciency by adjusting the full joint posterior for this modifed cross-validation, thus eliminating the need for model reftting. This method is implemented in the R-INLA package (www.r-inla.org) and can be adapted to a variety of inferential frameworks. |
| title | Leave-group-out cross-validation for latent Gaussian models |
| topic | Computation Applications Methodology 62-04 62C10 62F15 62J12 |
| url | https://arxiv.org/abs/2210.04482 |