The Time-Lagged Effect Problem on (Un)Truthful Data, a Case Study on Covid-19 Outbreak

The Coronavirus SARS-CoV-2 (COVID-19) emerged by December 2019, in Wuhan, China; it was reported and, a few months after, in the most of countries we are living a pandemic of an (almost) unknown disease never observed before. For this reason, the importance of a good measurement on the counting of observed cases has a crucial role. This work addresses the time-lagged effect problem via Bayesian analysis supported by a stochastic discrete-event simulation to give an answer to the truthfulness of the data and to validate the obtained results in terms of proportions based on an expected result in the particular case of Spain. Obtained results show that the reported data is untruthful and can make wrong any analysis, but even when the simulating results are as we expected they might be wrong in terms of absolute numbers. However, the most important knowledge we get is related to the fact that the disease might be considered under control because it is more likely that a person gets recover than She/He dies. © 2021, Springer Nature Switzerland AG.