Additive Penalized based Quantile Regression (APQR) for Predictive Analytics for Pandemic Data

ABSTRACT

In present, entire globe facing global challenges due to the pandemic situations raised by the known COVID-2019. Also, this can be impacted to most issues which can includes financial crises, medical emergency, education loss, chronic hunder, migration of people, etc. It can not only threat to the pre-existing people suffered with health issues also to the healthy people who are more conscious towards health. However, there is a necessity to apply a very good statistical analysis over such kind of issues. Moreover, many of the research works towards this problem entire globe. One of the popular techniques to apply statistical analysis to such kind of pandemic data is quantile regression. Need an extension version to the quantile regression to provide solution to the pandemic data analysis. Moreover, it is required to know prior the overall idea of the conditional distribution of a response variable. A penalized based quantile regression utilizes the minimization of the L1 norm to address heterogeneous pandemic data prediction. The study focused on minimization of L1 norm to effectively address such pandemic with additive penalized model for quantile regression known as APQR. The proposed model can help to find regression coefficients and control heterogeneity effectively. A numerical study using simulated and real examples demonstrates the competitive performance of the proposed APQR would be helpful and recommended for pandemic data prediction than standard quantile methods. © 2022, Mathematical and Research Society. All rights reserved.