Review of Forecasting Models for Coronavirus (COVID-19) Pandemic in India during Country-wise Lockdowns

Abstract

COVID-19 is spreading widely across the globe right now. Majority of the countries are relying on models and studies such as stochastic simulations, AceMod model, neural networks-based models, exponential growth model,Weibull distribution model, and so on to forecast the number of COVID-19 cases in the coming months. The objective on utilizing these models is to ensure that strict measures can be enacted to contain the virus spread and also predict the resources required to deal with the pandemic as the disease spreads. In the past few months, several models were used to predict the infection rate for COVID-19. These models predicted the infection rates, recovery rate or death rates for the COVID-19 patients. All these different models took different approaches and different scenarios to predict the future rates. Now, that we know the real cases, we can check how accurate these models were. Some of these models were able to predict the near future quite close to the reality but most of them went astray. In this study, we review major forecasting models that were used in the context of India during country-wise lockdown and compare them. From these comparisons, we can see that while the advanced warning can be helpful in mitigating and preparing for an impending or ongoing epidemic, effects can be catastrophic if poorly fitting models are used for predictions.

Fine-tuned Forecasting Techniques for COVID-19 Prediction in India

“…Building upon the previous research [18], current study implemented two numerical models to forecast the number of cases related to COVID-19 in India, namely – exponential curve fitting and least square fitted model. Both of the models forecasted an upward of 30 lakhs cases and 40,000 deaths for the upcoming months.…”

“…Short-term forecasting can be done based on elementary analytical approaches instead of diving into complex architectures like disease modelling or neural networks. Previous research [18] has shown that for shorter durations, simplistic curve fitting models achieve better accuracy than regression and pandemic models. Observing the patterns of number cases in countries such as China, Spain and Italy we can infer that the natural infection rate curve will follow a non-linear path initially till it hits its peak and begins to subside.…”

“…There has been a profusion of available mathematical techniques to predict the infection rate for the currently ongoing Covid-19 crisis. In past research [18], researchers evaluated the performance of majority of these techniques and concluded with two models which can be used for further purposes of estimating the number of cases affected by the coronavirus as these models gave the best predictions. These two models, exponential curve fitting and least square fitted model, can be used for short-term and long-term forecasting respectively.…”

Epidemic Models for COVID-19 during the First Wave from February to May 2020: a Methodological Review

“…Moreover, several of these reviews offer a categorization of the different models with a precise terminology [3,4,1] or a mapping of the articles with their key features [5,3]. Since November 2020, we can mention many projects of the latter type focusing on the COVID-19 epidemic, including published reviews [6,7,8] and continuously updated works of scientific watch [9,10].…”

A Quantitative Evaluation of COVID-19 Epidemiological Models

“…Some have focused on evaluating model performance on their weekly cumulative predictions. For instance, two different studies (5, 6) assessed the performance of different models with the median absolute percent error (MAPE) of cumulative deaths. Fried-man et al (5), observed that the calculated MAPE increased for longer forecasts, and the best performance model varied by region.…”