Logarithmic Quadratic Regression Model for Early Periods of COVID-19 Epidemic Count Data

Background: While COVID-19 epidemic has been spreading worldwide, its characteristics are still unclear. The development of good mathematical models for predicting prevalence and subsiding is strongly expected. curve shows how the increases subsides. This number persons found infected daily. To express this with a model, compartment model such as SIR used generally. However, parameter values these ordinary differential equation based very sensitive errors observed data, it often difficult to find reliable especially when amount data not sufficient. On other hand, regression small parameters more robust against than highly nonlinear though, clear what data.