Abstract
Coronavirus disease 2019 (COVID-19) is an infectious disease caused by severe acute respiratory syndrome coronvirus, which was declared as a global pandemic by the World Health Organization on March 11, 2020. In this work, we conduct a cross-sectional study to investigate how the infection fatality rate (IFR) of COVID-19 may be associated with possible geographical or demographical features of the infected population. We employ a multiple index model in combination with sliced inverse regression to facilitate the relationship between the IFR and possible risk factors. To select associated features for the infection fatality rate, we utilize an adaptive Lasso penalized sliced inverse regression method, which achieves variable selection and sufficient dimension reduction simultaneously with unimportant features removed automatically. We apply the proposed method to conduct a cross-sectional study for the COVID-19 data obtained from two time points of the outbreak.
Highlights
Since January 2020, many regions in China have experienced an outbreak of the coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus (SARS-Cov-2)
Motivated by Zou (2006) and Lin et al (2019), we propose an adaptive Lasso penalized sliced inverse regression method for the multiple index model to identify the possible risk factors for the infection fatality rate of COVID-19
Since the ALSIR method provides estimates of direction vectors for the basis of the central subspace derived from the multiple index model, positive or negative signs of estimates do not indicate positive or negative association of the covariate with the response
Summary
Since January 2020, many regions in China have experienced an outbreak of the coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus (SARS-Cov-2). It is critical to examine how the COVID-19 fatality may be associated with potential risk factors as a group instead of on an individual basis To this end, in this paper we employ the multiple index model to facilitate the relationship between possible risk factors and the fatality rate of COVID-19. Motivated by Zou (2006) and Lin et al (2019), we propose an adaptive Lasso penalized sliced inverse regression method for the multiple index model to identify the possible risk factors for the infection fatality rate of COVID-19. The proposed method develops a model-free variable selection procedure which does not require the specification of a parametric model for the underlying true process It estimates the central subspace of the multiple index model and selects the important features simultaneously.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
