A data driven approach for forecasting the COVID-19 cases in Pakistan

https://doi.org/10.5281/zenodo.17024984

Authors

  • Aasia Rajput Department of Statistics, Sindh Agriculture University, Tandojam, Pakistan
  • Naeem Ahmed Qureshi* Department of Statistics, University of Sindh, Jamshoro, Pakistan
  • Riaz Ali Buriro Department of Statistics, Sindh Agriculture University, Tandojam, Pakistan
  • Zeeshan Shabbir Soomro Department of Statistics, Sindh Agriculture University, Tandojam, Pakistan
  • Kainat Rajput Institute of Food Science and Technology, Sindh Agriculture University Tandojam, Pakistan

Abstract

Aimed at forecasting the COVID-19 cases in Pakistan using probability distribution modelling approach, the present study used daily data from 1st July to 30th September 2021 were used for estimation purpose whereas the data from 1st October to 31st December 2021 were used for forecasting. Based on the two modes and positively skewed behaviour of data, Bi-model lognormal distribution was fitted. The results showed that maximum number of daily new cases and recovered cases were recorded during the first wave while the maximum of daily deaths was observed during the third wave.  A retrospective analysis for evaluation of forecasts showed that the fitted model performed well leaving out-of-sample forecast error within the limit of 5% except for total recovered cases. Forecasted values for total cases and total deaths showed the decreasing pattern while increasing trend was observed for the total recovered cases.

Keywords: COVID-19; SARS-CoV-2; coronavirus; fifth wave; phenomenological epidemiologic model; bimodal distribution; lognormal distribution

Downloads

Published

2025-09-01

How to Cite

Aasia Rajput, Naeem Ahmed Qureshi*, Riaz Ali Buriro, Zeeshan Shabbir Soomro, & Kainat Rajput. (2025). A data driven approach for forecasting the COVID-19 cases in Pakistan: https://doi.org/10.5281/zenodo.17024984. `, 4(01), 3332–3344. Retrieved from https://assajournal.com/index.php/36/article/view/816

Issue

Vol. 4 No. 01 (2025): Advance Social Science Archive Journal

Section

Articles

Most read articles by the same author(s)