Mathematical forecasts are hypotheses: considering uncertainty in publishing COVID-19 pandemic data

Abstract

The current period is, at the same time, highly uncertainty and intensive in data production. In the analysis of scenarios produced by the COVID-19 pandemic, we make forecast based on mathematical models where exact assumptions are never true. This leads to a reconsideration of the usual form of data analysis, both in journalism and academic research, bolding on the need for exhaustive testing of the available information, since researchers are often not clear enough in their parameters or assumptions. These omissions usually become in different levels of importance in the media, because journalists tend to assume the same attitudes towards complex data just in the way they do with the news in "normal periods". However, it is necessary to obtain an adequate understanding of the data on positive COVID-19 cases, defunctions or politics aimed to containing the virus, designing some basic rules that both specialists and journalists may take into account when they have to write papers or publish news at a time as critical as nowadays.