Covid-19 Study

Infected cases forecast model: SIR (Susceptible-Infective-Recovered) model is used to forecast the number of infected cases, and the peak time.
A model of reactive social distancing in epidemics is proposed, in which the infection rate changes with the number infected. The final-size equation for the total number that the epidemic will infect can be derived analytically, as can the peak infection proportion. This model could assist planners during for example the COVID-19 pandemic.

  • We solve required differential equations by using Python and fit the data curve.
  • We first estimate the parameters. We focus on the reproduction rate r -- assumption here is close to 2.
  • Social distancing is considered to flatten the curve, that is slowing the rate of infection.
  • Parameter gamma (g) regulates social distancing impact.

The following graphs illustrate the infection case numbers are given in proportions to populations. Please note the days are from the day after the confirmed 100th case. They show the best possible forecasts and the average as well as the estimated peak date and the maximum infection proportion.

April 14, 2020

April 13, 2020

April 11, 2020

April 7, 2020

April 4, 2020

April 3, 2020

April 2, 2020

We updated forecasts for New York and for all US other than New York. The peak day shows for New York as mid April and for the other places are expected to be lagging 2-3 weeks. Latest forecasts are below; please note that these are not total numbers that we expect to see when the epidemic is over. We expect to have total confirmed cases is about 4.760.000 and total fatality is about 95.600. These numbers are close to recently published White House projections.

April 1, 2020

March 31, 2020

We use ideas from, check them out: here

Our Expert

  • Dr. Cetin Urtis, Research Scientist ▼


    Dr. Urtis - PhD from University of Minnesota.
    Thesis Title: Special Values of L-functions by a Siegel-Weil-Kudla-Rallis Formula

    • Developed SIR (Susceptible-Infective-Recovered) model to predict the number infected Covid19 cases, and the peak time.
    • Developed a recession probability model that estimated the probability of the start of a recession over the next 12 months using macroeconomic indicators, and Logistic and Probit regression models
    • Conducted research on “Spectral Theory of Automorphic Forms” which involves functional analysis, measure theory (related to SVM reproducing kernel) and harmonic analysis (related to CNN-convolutional neural networks)

    Enjoy solving complex and challenging problems using mathematical tools. Expert in extracting business insights, developing technical ideas and communicating results to a nontechnical or executive audience.

    • The Fulbright Visiting Scholar Program Grant
    • University of Minnesota “Citation for Excellence in Teaching”
    • NATO-A1 Fellowship
    • Honorable Mention, 32nd International Mathematical Olympiad, Sigtuna, Sweden

Our Solutions

We believe analytics isn’t something you can buy; it requires planning, coordination and a vision. The following core areas of intelligent business must be integrated to create the smarter enterprise!