| 1. |
- Sherratt, K., et al.
(författare)
-
Predictive performance of multi-model ensemble forecasts of COVID-19 across European nations
- 2023
-
Ingår i: eLIFE. - : eLife Sciences Publications Ltd. - 2050-084X. ; 12
-
Tidskriftsartikel (refereegranskat)abstract
- Methods: We used open-source tools to develop a public European COVID-19 Forecast Hub. We invited groups globally to contribute weekly forecasts for COVID-19 cases and deaths reported by a standardised source for 32 countries over the next 1–4 weeks. Teams submitted forecasts from March 2021 using standardised quantiles of the predictive distribution. Each week we created an ensemble forecast, where each predictive quantile was calculated as the equally-weighted average (initially the mean and then from 26th July the median) of all individual models’ predictive quantiles. We measured the performance of each model using the relative Weighted Interval Score (WIS), comparing models’ forecast accuracy relative to all other models. We retrospectively explored alternative methods for ensemble forecasts, including weighted averages based on models’ past predictive performance.Results: Over 52 weeks, we collected forecasts from 48 unique models. We evaluated 29 models’ forecast scores in comparison to the ensemble model. We found a weekly ensemble had a consistently strong performance across countries over time. Across all horizons and locations, the ensemble performed better on relative WIS than 83% of participating models’ forecasts of incident cases (with a total N=886 predictions from 23 unique models), and 91% of participating models’ forecasts of deaths (N=763 predictions from 20 models). Across a 1–4 week time horizon, ensemble performance declined with longer forecast periods when forecasting cases, but remained stable over 4 weeks for incident death forecasts. In every forecast across 32 countries, the ensemble outperformed most contributing models when forecasting either cases or deaths, frequently outperforming all of its individual component models. Among several choices of ensemble methods we found that the most influential and best choice was to use a median average of models instead of using the mean, regardless of methods of weighting component forecast models.Conclusions: Our results support the use of combining forecasts from individual models into an ensemble in order to improve predictive performance across epidemiological targets and populations during infectious disease epidemics. Our findings further suggest that median ensemble methods yield better predictive performance more than ones based on means. Our findings also highlight that forecast consumers should place more weight on incident death forecasts than incident case forecasts at forecast horizons greater than 2 weeks.
|
|
| 2. |
- Teixeira, J., et al.
(författare)
-
Tropical and Subtropical Cloud Transitions in Weather and Climate Prediction Models : The GCSS/WGNE Pacific Cross-Section Intercomparison (GPCI)
- 2011
-
Ingår i: Journal of Climate. - 0894-8755 .- 1520-0442. ; 24:20, s. 5223-5256
-
Tidskriftsartikel (refereegranskat)abstract
- A model evaluation approach is proposed in which weather and climate prediction models are analyzed along a Pacific Ocean cross section, from the stratocumulus regions off the coast of California, across the shallow convection dominated trade winds, to the deep convection regions of the ITCZ-the Global Energy and Water Cycle Experiment Cloud System Study/Working Group on Numerical Experimentation (GCSS/WGNE) Pacific Cross-Section Intercomparison (GPCI). The main goal of GPCI is to evaluate and help understand and improve the representation of tropical and subtropical cloud processes in weather and climate prediction models. In this paper, a detailed analysis of cloud regime transitions along the cross section from the subtropics to the tropics for the season June-July-August of 1998 is presented. This GPCI study confirms many of the typical weather and climate prediction model problems in the representation of clouds: underestimation of clouds in the stratocumulus regime by most models with the corresponding consequences in terms of shortwave radiation biases; overestimation of clouds by the 40-yr ECMWF Re-Analysis (ERA-40) in the deep tropics (in particular) with the corresponding impact in the outgoing longwave radiation; large spread between the different models in terms of cloud cover, liquid water path and shortwave radiation; significant differences between the models in terms of vertical cross sections of cloud properties (in particular), vertical velocity, and relative humidity. An alternative analysis of cloud cover mean statistics is proposed where sharp gradients in cloud cover along the GPCI transect are taken into account. This analysis shows that the negative cloud bias of some models and ERA-40 in the stratocumulus regions [as compared to the first International Satellite Cloud Climatology Project (ISCCP)] is associated not only with lower values of cloud cover in these regimes, but also with a stratocumulus-to-cumulus transition that occurs too early along the trade wind Lagrangian trajectory. Histograms of cloud cover along the cross section differ significantly between models. Some models exhibit a quasi-bimodal structure with cloud cover being either very large (close to 100%) or very small, while other models show a more continuous transition. The ISCCP observations suggest that reality is in-between these two extreme examples. These different patterns reflect the diverse nature of the cloud, boundary layer, and convection parameterizations in the participating weather and climate prediction models.
|
|
| 3. |
- Chavanne, A. V., et al.
(författare)
-
Individual-Level Prediction of Exposure Therapy Outcome Using Structural and Functional MRI Data in Spider Phobia: A Machine-Learning Study
- 2023
-
Ingår i: Depression and Anxiety. - 1091-4269. ; 2023
-
Tidskriftsartikel (refereegranskat)abstract
- Machine-learning prediction studies have shown potential to inform treatment stratification, but recent efforts to predict psychotherapy outcomes with clinical routine data have only resulted in moderate prediction accuracies. Neuroimaging data showed promise to predict treatment outcome, but previous prediction attempts have been exploratory and reported small clinical sample sizes. Herein, we aimed to examine the incremental predictive value of neuroimaging data in contrast to clinical and demographic data alone (for which results were previously published), using a two-level multimodal ensemble machine-learning strategy. We used pretreatment structural and task-based fMRI data to predict virtual reality exposure therapy outcome in a bicentric sample of N=190 patients with spider phobia. First, eight 1st-level random forest classifications were conducted using separate data modalities (clinical questionnaire scores and sociodemographic data, cortical thickness and gray matter volumes, functional activation, connectivity, connectivity-derived graph metrics, and BOLD signal variance). Then, the resulting predictions were used to train a 2nd-level classifier that produced a final prediction. No 1st-level or 2nd-level classifier performed above chance level except BOLD signal variance, which showed potential as a contributor to higher-level prediction from multiple regions across the brain (1st-level balanced accuracy=0.63). Overall, neuroimaging data did not provide any incremental accuracy for treatment outcome prediction in patients with spider phobia with respect to clinical and sociodemographic data alone. Thus, we advise caution in the interpretation of prediction performances from small-scale, single-site patient samples. Larger multimodal datasets are needed to further investigate individual-level neuroimaging predictors of therapy response in anxiety disorders.
|
|
| 4. |
|
|
| 5. |
|
|
| 6. |
|
|
