Mathematical modeling is one of the widely used scienti c tools which allow us to predict the evolution of the disease and take intervention measures accordingly. At least four types of models can be used to study the evolution of infectious diseases: the compartmental models (deterministic di erential equations like SIR or SEIR ), agent-based models (that consider people as lattice sites on a network, each site being in a speci c stage-susceptible, exposed, infected, etc.- that is modi ed according to some precise rules), stochastic di erential equations (di erential equations which feature random variables) and data-driven model (that simply take the existing data disease's spread over a period and use machine learning methods to generate a forecast for the next short period, paying little or no attention to the underlying processes driving the spread of the disease). Each of them has advantages and limitations. The philosophy of the grey modeling is the use of appropriate deterministic models (dynamical systems) whose coecients are determined using data series from measurements. It is a combination of the compartmental and data-driving techniques. The purpose of the article is to apply the grey modeling in order to identify a mathematical model that ts as accurately as possible the statistical data on the spread of COVID 19 in Romania. The study considers two categories of population (infected, respectively vaccinated people) about which daily statistical data are available. We study each population using the grey Verhulst model and we try to point out a correlation between them using the grey Lotka-Volterra model. Short time predictions are also presented and discussed.