论文标题
关于使用众包数据的冠状病毒疾病爆发的早期流行病学分析的注释
A Note on Early Epidemiological Analysis of Coronavirus Disease 2019 Outbreak using Crowdsourced Data
论文作者
论文摘要
众包数据可以证明在监测和控制传染病的传播方面至关重要。 Sun，Chen和Viboud（2020）最近发表的论文很重要，因为它有助于理解流行病学和在大多数流行特征仍然未知的时期中对Covid-19的传播。但是，众包数据的使用从统计的角度引起了许多问题，这些观点的风险冒着结果无效的风险以及偏见估计和假设检验的风险。鉴于该主题对全球卫生安全的重要性，必须赞扬Sun，Chen和Viboud（2020）的作品，但在本文中，我们认为重要的是要指出可能存在统计偏见的来源，并指出可能向他们指出可能的解决方案。
Crowdsourcing data can prove of paramount importance in monitoring and controlling the spread of infectious diseases. The recent paper by Sun, Chen and Viboud (2020) is important because it contributes to the understanding of the epidemiology and of the spreading of Covid-19 in a period when most of the epidemic characteristics are still unknown. However, the use of crowdsourcing data raises a number of problems from the statistical point of view which run the risk of invalidating the results and of biasing estimation and hypothesis testing. While the work by Sun, Chen and Viboud (2020) has to be commended, given the importance of the topic for worldwide health security, in this paper we deem important to remark the presence of the possible sources of statistical biases and to point out possible solutions to them
