The purpose of this paper is to develop the fractional stochastic nonlinear differential equation approach (FBM) to apply this in the study of the evolution in time of infected number by coronavirus in countries where the number of infected is large. The rises and falls of epidemiologic weeks of novel cases are treated as fractional white noise BH (t); 1/2 < H < 1, in the model for spreading of epidemiologic week of novel cases. Since the FBM approach provide a useful model for a host of natural time series presented by scientists, engineers and statisticians, we apply this formalism to study the spreading of coronavirus. Moreover, we use the rescaled range analysis (RS) to determine the Hurst index of the time series of epidemiologic weeks of novel cases which exhibits a Hurst index much larger than 1/2, what means that the time series has a very distinct behavior from a random walk or that the nth value of time series up time t 0 is independent of all the values before, t < t 0.