Modeling of Total Cases due to COVID-19 and its Impact in India

Abstract

This research paper focuses on the modeling of total cases due to COVID-19 and the critical assessment of socioeconomic impact on India. The data set considered for the present analysis is from December 31, 2019 to May 16, 2020 for training and testing of developed regression model. Least-square approximation of linear regression technique is applied to estimate the total cases of COVID-19. Three variables, viz. daily new cases, total deaths and daily new deaths, were considered for development of correlations. In the present study, seven correlations are developed as a function of single variable, two variables and three variables with accuracy (R2) ranging from 85.71 to 99.95%. The paper also highlights the socioeconomic impact of COVID 19 on different sector, challenges and remedies for improving the GDP of the country.

Introduction

In nineteenth century, the coronavirus has been primarily associated with non-severe respiratory infections in human beings. Afterward, three different types of coronavirus have combined and widespread across the several countries [1,2,3]. At the beginning of the twentieth century, the severe acute respiratory syndrome (SAR) caused by coronavirus was diagnosed in China and spread in 29 countries with 8096 confirmed cases and 774 deaths [3,4,5,6]. In September 2012, the first case of Middle East respiratory syndrome (MERs) coronavirus was reported with 2458 confirmed cases and 848 deaths in 27 countries [7,8,9,10].

In December 2019, a new coronavirus was isolated in patients related to a fish market in the city of Wuhan, Hubei Provence of China. This coronavirus named SARs-CoV-2 caused an epidemic in that city that has spread rapidly in the world, possibly the largest pandemic after the Spanish flu [11,12,13]. As the numbers of patients of pneumonia were increasingly tremendously in the Wuhan, the municipal health commission reported a cluster of cases of pneumonia while considering it as an epidemic healthcare issue which arises from pneumonia [14,15,16]. A novel coronavirus was eventually identified. Due to human-to-human transmission, the total cases of coronavirus in the world are around 4,621,409 on May 16, 2020 and more than 75% of total cases of the world shared by these countries with their percentage sharing as USA (32.12%), Spain (5.94%), Russia (5.69%), UK (5.12%), Italy (4.84%), Brazil (4.72%), France (3.88%), Germany (3.80%), Turkey (3.17%), Iran (2.52%), India (1.86%) and China (1.79%) [16]. Total twenty worse epidemics and pandemics were studied under the critical assessment and enlisted in chronological order in Table 1.

Table 1 Chronological history of worse epidemics and pandemics [31]

WHO issued worldwide health emergency, indicating that SARs-CoV-2 is of urgent global concerned and estimated that reproduction factor (R0) as 2.7. WHO time to time had performed its task during the critical situation and declared public health emergency of international concern (PHEIC) sixth time after the international health regulations came into existence in 2005. On January 30, 2020, WHO reported 7818 total confirmed cases worldwide with majority of these in China and 82 cases reported in 18 countries across the globe. Figure 1 illustrates the details of time-line plan of WHOs from December 31, 2019 to March 18, 2020. Presently, an international clinical trial is generating data from all medical organizations including WHO and researchers are collecting data on coronavirus throughout the world to find the most effective treatments for COVID-19 [16,17,18].

Fig. 1
figure1

WHO’s time line of COVID-19 [16]

Numerous statistical techniques have gained momentum and are playing important role in the epidemiological data analysis. Statistical technique also can be used to develop standard mortality models like machine learning (ML) [19], artificial neural network (ANN) [20], probabilistic neural network (PNN) [21], Gaussian process for regression (GRNN) [22], auto-regressive integrated moving average (ARIMA) [23] and dynamic modeling [24]. WHO has given some guidelines and recommendations for handling of coronavirus patients. As per the recommendations, Equation A is used to calculate the active cases confirmed and affected by corona virus,

$$Active Cases=Total Cases-Total Deaths-Recovered$$
(1)

where total cases can be calculated as the addition of confirmed positive, presumptive and suspect or probable cases. Total deaths can be calculated as cumulative number of deaths among detected cases. The recovered can be calculated as count of total patients discharged from hospitals, but here in recovered cases, some of the consequences were observed as imperfection in data submission from state to country level or local to national levels and vice versa. To calculate the recovered cases, the WHO recommends the following criteria of [symptoms resolve + 2 negative tests within 24 h] or [symptoms resolve + additional 14 days], but this is only recommendations [16]. In line with the relevant literature, the present work aimed to propose regression correlations for predicting total cases of COVID-19. The result of the study may facilitate the public health officials of the Indian government for prevention and control measures of COVID-19.

Social Impact Due to COVID-19

The COVID-19 pandemic affected the educational systems such as schools, colleges and universities worldwide due to its widespread. According to the UNESCO report published on 25 March 2020, around 165 out of 195 countries of the world had closed educational systems due to COVID-19. Because of widespread of COVID-19, an individual with low income suffered mostly and die quickly. It has been observed that the COVID-19 spreads very quickly in crowded residential, markets and workers in low skill jobs [3, 7, 8].

Economic Impact Due to COVID-19

WHO declared COVID-19 as a pandemic in January 2020, and it has spread in the worldwide as like wildfire. In the initial stage, most of the countries were taking COVID-19 as a cluster effect of pneumonia, but its adverse effect was observed in human-to-human transmission. Many of the countries have spent money on the development of medical infrastructures and medicines related to COVID-19. Due to pandemic adverse effect, all the countries across the globe suffered GDP loss. According to International Monitory Fund (IMF), world economic outlook report of April 2020, countries like USA, Germany, France, Italy and Spain had GDP loss ranging from − 5.9 to − 9.1%, whereas India and China have GDP loss ranging from 1.9 to 1.2%. As per the projection of financial year 2021, all the countries mentioned in Fig. 2 will be coming up in positive GDP [25, 26].

Fig. 2
figure2

Countries with their actual and projected GDP’s [25, 26]

Socioeconomic Impact and Challenges

Understanding the turmoil effect of the economy is of paramount importance due to COVID 19. It is equally important to summarize the socioeconomic effect of various sectors that include primary sector (agriculture, petroleum and oil) basically involved in extraction of raw material, secondary sector (manufacturing industry), tertiary sectors (education, finance sector, health care and pharmaceutical, tourism, hospitality, aviation, real estate, housing, sports, IT, media food, etc.) and other impacts are social impacts (family, society, etc.). Few of the challenges in front of Indian government are building of medical care facility at macro and micro-level, continuity in supply chain management in commodities, effective utilization of resources, sanitization in containment zone and other areas and issue associated with migrant workers.

On March 25, government invoked Disaster Mitigation Act 2005 and imposed national lockdown and extended it in phases, cancellation of flights at domestic and international level, suspension of train, bus and metro services, sealing of state borders, dedicated COVID-19 quarantine sites, extension of certain compliance deadlines, launching of other IT application like Aarogya Setu App. Recently, government of India has declared rupees 20 lakh crore relief package to boost different sectors such as small/medium enterprises including MSMEs, NBFs/HFCs/MFIs also reduced the TDS and TCS rates, free food grain supply to migrant workers for 2 months, special credit facility to street vendors, housing CLSS-MIG, additional credit through KCC, Agri-infrastructure fund, additional MGNREGS allocations, etc. The fund was allocated by government of India in five parts in between March 2020 to mid of the May 2020. The Economist Intelligence Unit has forecasted the GDP growth rate for 2020–21 for India at 2.1% when compared to China and USA as 1% and −2.8% respectively [27, 28].

Results and Discussion

In this section, the results of statistical techniques applied in the present study are presented along with the interrelated recommendations for the improvement of GDP after COVID-19.

Numerical Modeling

Least-square approximation of linear regression technique has been utilized in the calculation of total cases of COVID-19 pandemic. The total cases can be calculated as a function of daily new cases, total deaths and daily new death's. Different regression equations are developed from the correlational analysis.

$${{\varvec{T}}{\varvec{o}}{\varvec{t}}{\varvec{a}}{\varvec{l}}\boldsymbol{ }{\varvec{C}}{\varvec{a}}{\varvec{s}}{\varvec{e}}{\varvec{s}}}_{{\varvec{E}}{\varvec{s}}{\varvec{t}}.}=f(Daily new cases, Total deaths, Daily new deaths)$$
(2)

For the analysis of COVID-19 statistical data of actual values of total cases, new cases, total deaths and new deaths are taken from WHO website from December 31, 2019 to May 16, 2020 [29]. Figure 3 illustrates the information of actual values of active cases and deaths due to COVID-19. To simplify the analysis, total deaths were measured in thousands and daily new deaths measured in numbers and dates were shown in step of ten days for simplicity. The total numbers of cases in India were less than 1100 up to March 30, 2020 and increased at an exponential rate after April 1, 2020 as shown in Fig. 3. Total cases are calculated by using seven different combinations; Carl Pearson coefficient (R2) and constants obtained from the regression analysis are depicted in Table 2. The R2 value of the developed correlations is in the range of 85.71–95%. The correlations developed by using single, two and three variables under considerations are shown below:

Fig. 3
figure3

Actual values of active cases and deaths due to COVID-19 [29]

Table 2 Constants of Regression equation
$${Total Cases}_{Est.}=-1062+17.554*Daily New Cases$$
(3)
$${Total Cases}_{Est.}=109.8+30.5211*Total Deaths$$
(4)
$${Total Cases}_{Est.}= -225+506.5*Daily New Deaths$$
(5)
$${Total Cases}_{Est.}=126-0.177*Daily New Cases+30.815*Total Deaths$$
(6)
$${Total Cases}_{Est.}=-1035+20.48*Daily New Cases-92.7*Daily New Deaths$$
(7)
$${Total Cases}_{Est.}=152.5+31.179*Total Deaths-12.68*Daily New Deaths$$
(8)
$${Total Cases}_{Est.}=122+30.564*Total Deaths+0.584*Daily New Cases-19.54*Daily New Deaths$$
(9)

Table 3 illustrates the percentile error observed in estimated and actual values for the total cases. The highlighted cell shows the highest and the lowest percentile error given by the regressions correlations.

Table 3 Percentile error observed in estimated values from actuals

Figure 4 shows the estimated values obtained from the developed correlations compared with the actual total cases estimated by WHOs [25].

Fig. 4
figure4

Comparison of actual and estimated values of total cases

Minimum and maximum percentile observed in developed correlations ranges from 0.002 to 4.171. These correlations can also be useful for predicting data at macro-level (country and state) micro-level (district, cities, municipal corporation, villages, etc.). In the present study, the data set under consideration is from December 31, 2019 to May 16, 2020. However, the developed correlations are applied for predicting the total cases for May 17, 2020 with the deviation of ± 3%.

Remedies for Improving GDP

Identifying the effective response strategies to fight against COVID-19 is very important. Eminent scientist, doctors and health officials are trying to identify preventive and treatment strategies to defeat COVID-19. Epidemiological modeling needs different approach as compare to engineering problem. The author believes that the present study will help the public health officials to handle COVID-19 problem which is unstable and open ended in nature. Principal of control theory involving feedback system can be applied to normalize the issue of COVID-19 with various approaches to manage trade-off between economy and health outcome. The International Monitory Fund (IMF) has cut the India’s growth forecast for 2020–21 to 1.9% down from its earlier estimate of 5.8% in January this year. Also workers and members of lower income group have been hit hard as their wages disappeared. The International Labour Organization (ILO) estimates that 400 million peoples in India are at risk of sinking deeper into poverty [30]. There is a strong relationship between social, health, environment and economy. As India is trying to boost up its economy, exploring new resources of growth and identifying systematic actions are needed for sustainable economic growth. There are few interrelated recommendations to boost economy [31].

  • Investment on cleaner production technologies.

  • Application of virtual technology such as information communication technology (ICT) tools in all sectors.

  • Online and digitization in various sectors.

  • Relook and revamp to take into an account new drivers of growth in biotechnology and pharmaceuticals, renewable energy and organic farming.

  • Expand excess to clean water and air.

  • Application of fiscal mechanism.

  • Promotion of self-sufficiency of economics in health, education, agriculture and communications.

Limitations

Limitations of our current work are as follows:

  • Only linear relationship has been developed for the given data set.

  • Present study is only applicable for Indian context.

  • The study is limited to only short-term prediction.

Conclusions

COVID-19 has spread worldwide within a short span of time and affected a social and economic life of the human beings. WHO issued worldwide health emergency, indicating that SARs-CoV-2 is of urgent global concerned and estimated that reproduction factor (R0) as 2.7. This research paper proposed a least-square approximation of linear regression technique for estimating the total cases of COVID-19 in India as a function of daily new cases, total deaths and daily new deaths. Seven correlations were developed to predict the total cases with the average accuracy of 96.65%. This kind of study is capable of providing a predictive solution to current and future pandemics and can be useful for public health officials for taking better decisions. Due to interrelationship between social, health, environment and economy, the country is vulnerable to economic downturn and needs greater access to new sources of growth that will improve life expectancy and economy. In the future, the study will be enhanced by adding variables like temperature and oxygen level of the patients.

Abbreviations

%:

Percentage

AD:

Anno Domini

BC:

Before Christ

GDP:

Gross domestic products

IMF:

International monitory fund

k:

Numbers in thousands

MERs:

Middle East respiratory syndrome

PHEIC:

Public health emergency of international concern

R2 :

Carl Pearson coefficient

SARs:

Severe acute respiratory syndrome

WHO:

World health organization

TDS:

Tax deducted at source

TCS:

Tax collected at source

MSMEs:

Ministry of micro, small & medium enterprises

NBFs:

Non-banking finance companies

HFCs:

Housing finance companies

MFIs:

Merchandise exports from India scheme

CLSS-MIG:

Credit-linked subsidy scheme (CLSS) for middle-income groups (MIG)

KCC:

Kisan Credit Card

MGNREGS:

Mahatma Gandhi National Rural Employment Guarantee Act

References

  1. 1.

    I. Abdeljaoued-Tej, COVID-19 data analysis and modeling in Palestine. medRxiv 9(3), 50 (2020)

  2. 2.

    T. Chakraborty, I. Ghosh, Real-time forecasts and risk assessment of novel coronavirus (COVID-19) cases: a data-driven analysis. Chaos Solitons Fractals 135, 109850 (2020)

    Article  Google Scholar 

  3. 3.

    K. Chatterjee, K. Chatterjee, A. Kumar, S. Shankar, Healthcare impact of COVID-19 epidemic in India: a stochastic mathematical model. Med. J. Armed Forces India 76(2), 147–155 (2020)

    Article  Google Scholar 

  4. 4.

    P. Chatterjee, Indian pharma threatened by COVID-19 shutdowns in China. The Lancet 395(10225), 675 (2020)

    Article  Google Scholar 

  5. 5.

    N. Chintalapudi, G. Battineni, F. Amenta, COVID-19 disease outbreak forecasting of registered and recovered cases after sixty day lockdown in Italy: a data driven model approach. J. Microbiol. Immunol. Infect. 53(3), 396–403 (2020)

    Article  Google Scholar 

  6. 6.

    H.G. Garcell, COVID-19. Un reto para los profesionales de la salud. Revista Habanera de Ciencias Médicas 19(2), 3284 (2020)

    Google Scholar 

  7. 7.

    Q. Li, W. Feng, Y.H. Quan, Trend and forecasting of the COVID-19 outbreak in China. J. Infect. 80(4), 469–496 (2020)

    Google Scholar 

  8. 8.

    R. Li, S. Qiao, G. Zhang, Analysis of angiotensin-converting enzyme 2 (ACE2) from different species sheds some light on cross-species receptor usage of a novel coronavirus 2019-nCoV. J. Infect. 80(4), 469–496 (2020)

    Google Scholar 

  9. 9.

    Y.S. Long, Z. M. Zhai, L.L. Han, J. Kang, Y.L. Li, Z.H. Lin, Z. Liu, Quantitative assessment of the role of undocumented infection in the 2019 novel coronavirus (COVID-19) pandemic. arXiv preprint (2020). https://arxiv.org/abs/2003.12028

  10. 10.

    S. Mittal, An exploratory data analysis of COVID-19 in India. Int. J. Eng. Res. Technol. (IJERT) (2020). https://doi.org/10.17577/IJERTV9IS040550

    Article  Google Scholar 

  11. 11.

    G. Pandey, P. Chaudhary, R. Gupta, S. Pal (2020). SEIR and regression model based COVID-19 outbreak predictions in India. arXiv preprint https://arxiv.org/abs/2004.00958

  12. 12.

    M. Perc, N. Gorišek Miksić, M. Slavinec, A. Stožer, Forecasting Covid-19. Front. Phys. 8, 127 (2020)

    Article  Google Scholar 

  13. 13.

    F. Petropoulos, S. Makridakis, Forecasting the novel coronavirus COVID-19. PLoS ONE 15(3), e0231236 (2020)

    Article  Google Scholar 

  14. 14.

    R. Singh, R. Adhikari, Age-structured impact of social distancing on the COVID-19 epidemic in India. arXiv preprint (2020). https://arxiv.org/abs/2003.12055

  15. 15.

    T. Singhal, A review of coronavirus disease-2019 (COVID-19). Indian J. Pediatr. 87(4), 281–286 (2020)

    Article  Google Scholar 

  16. 16.

    https://www.who.int/news-room/detail/27-04-2020-who-timeline---covid-19 Date 16 May 2020 IST 3.33 pm

  17. 17.

    C.N. Van Dijk, Covid-19. J. ISAKOS 5(3), 115–116 (2020)

    MathSciNet  Article  Google Scholar 

  18. 18.

    S. Zhang, M. Diao, W. Yu, L. Pei, Z. Lin, D. Chen, Estimation of the reproductive number of novel coronavirus (COVID-19) and the probable outbreak size on the Diamond Princess cruise ship: a data-driven analysis. Int. J. Infect. Dis. 93, 201–204 (2020)

    Article  Google Scholar 

  19. 19.

    P. Deprez, P.V. Shevchenko, M.V. Wüthrich, Machine learning techniques for mortality modeling. Eur. Actuar. J. 7(2), 337–352 (2017)

    MathSciNet  Article  Google Scholar 

  20. 20.

    D. Hainaut, A neural-network analyzer for mortality forecast. ASTIN Bull. J. IAA 48(2), 481–508 (2018)

    MathSciNet  Article  Google Scholar 

  21. 21.

    R. Richman, M.V. Wüthrich, A neural network extension of the Lee-Carter model to multiple populations. Ann. Actuar. Sci. 7, 1–21 (2018)

    Google Scholar 

  22. 22.

    M. Ebden (2008). Gaussian processes for regression: a quick introduction. [online] Available at:< http://www.robots.ox.ac.uk/~ mebden/reports. GPtutorial. pdf

  23. 23.

    A. Khamparia, A. Singh, D. Anand, D. Gupta, A. Khanna, N.A. Kumar, J. Tan, A novel deep learning-based multi-model ensemble method for the prediction of neuromuscular disorders. Neural Comput. Appl. 32(15), 11083–11095 (2018)

    Article  Google Scholar 

  24. 24.

    S. Karthick, K.S. Kumar, S. Mohan, Relative analysis of controller effectiveness for vertical plane control of an autonomous underwater vehicle. OCEANS 2016 -Shanghai IEEE (2016). https://doi.org/10.1109/OCEANSAP.2016.7485569

    Article  Google Scholar 

  25. 25.

    http://statisticstimes.com/economy/projected-world-gdp-ranking.php Date: 16 May 2020 IST 12.10 am

  26. 26.

    https://www.imf.org/en/About/Factsheets/IMF-at-a-Glance Date 16 May 2020 IST 6.08 pm

  27. 27.

    https://en.wikipedia.org/wiki/Social_impact_of_the_COVID-19_pandemic Date: 16 May 2020 IST 12.13 am

  28. 28.

    https://www.india-briefing.com

  29. 29.

    https://ourworldindata.org/coronavirus-source-data date: 16 May 2020 IST 12.04 am

  30. 30.

    https://www.wri.org/blog/2020/04/after-covid-19-5-ways-india-can-pursue-sustainable-and-resilient-recovery Date: 26 October 2020 IST 4.27 pm

  31. 31.

    https://www.livescience.com/worst-epidemics-and-pandemics-in-history.html Date 16 May 2020 IST 5.50 pm

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Siraj Sayyed.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

See Fig. 

Fig. 5
figure5

Flow chart of modeling

5.

See Table 4.

Table 4 Estimation of total cases through regression analysis and calculation of absolute error

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kulkarni, K., Kulkarni, A., Shaikh, N.S. et al. Modeling of Total Cases due to COVID-19 and its Impact in India. J. Inst. Eng. India Ser. B (2021). https://doi.org/10.1007/s40031-021-00558-w

Download citation

Keywords

  • Coronavirus
  • COVID-19
  • Epidemics
  • Pandemics
  • Modeling
  • Socioeconomic impact