Time Series analysis is being widely used in the fields of corporate sectors, medical sciences, weather forecasting and stock prices to predict future values based on previously observed values. In this research, Time Series data of COVID-19 (SARS-CoV-2) obtained from a dataset obtained from worldometers has been to predict the prognosis of the virus in the forthcoming days. The dataset contains summed up values of the daily new cases, recovered cases and deaths from 187 countries right from the day from which the first case was reported. After exploring and splitting the data into train and test sets to ensure the fidelity of the model, a Machine Learning model called Prophet which was introduced by Facebook has been used to train and test on the observed data. Finally, it has been used to draw probabilistic insights for the future on the new cases, recovered cases and new deaths for the next month.