Prediction model for rare events in longitudinal follow-up and resampling methods
This work addresses the challenge of rare event prediction in longitudinal studies, which is important for fields like healthcare or epidemiology, but it appears incremental as it focuses on comparing existing resampling methods.
The authors tackled the problem of building prediction models for rare events in longitudinal follow-up studies by comparing resampling methods to improve standard regression models, evaluating the effect of sampling rate on predictive performance using a real-life example and a validation technique that accounts for time.
We consider the problem of model building for rare events prediction in longitudinal follow-up studies. In this paper, we compare several resampling methods to improve standard regression models on a real life example. We evaluate the effect of the sampling rate on the predictive performances of the models. To evaluate the predictive performance of a longitudinal model, we consider a validation technique that takes into account time and corresponds to the actual use in real life.