Longitudinal Impact of the Anticipated Outbreak of COVID-19 Using Logistic Growth Regression and Shanks Transformation

Abstract

This study uses a logistic regression growth (LRG) model to forecast the ultimate epidemic size caused by COVID-19 in different parts of Pakistan, such as Gilgit Baltistan and Islamabad. Nonlinear fits to the data are used to determine the model parameters. The locations where the curves show signs of flattening are ones where the logistic regression growth (LRG) model excels at providing comparative estimates. To account for this sustained drop, Pakistani data are adjusted. The number of examples is more promising when the Shanks transformation is iterated. For the areas most affected by COVID-19, these models provide estimates. These estimates suggest that in the United States, Italy, Spain, and Germany, 1.1, 0.31, 0.35, and 0.19 million people will be affected by the epidemic in the long run. Whether or not COVID-19 predictions can be made using a regression growth model and the Shanks transformation. The simplicity and speed with which this technique may be implemented make it a valuable resource for public health professionals and policymakers combating the COVID-19 pandemic. In addition, the curves of the second wave are forecast to flatten in most places by the middle of January 2020.