As time evolves, a community in a social network may experience different changes known as critical events. For example, a community can either split into a few different communities, project into a larger community, shrink to a smaller community, remain stable or merge into another network. Prediction of critical events has attracted in expanding attention in the ongoing writing. Learning in the advancement of communities after some time is a key advance towards predicting the critical events the communities may undergo. This is an essential and difficult issue in the study of social networks. In the work to date, there is a lack of formal methodologies for displaying and predicting basic events over time. This motivates our effort to plan another statistical technique for event expectation keeping in mind the end goal to improve utilization of histories of past changes. To this end, this project proposes a sliding window investigation from which we build up a model that at the same time abuses an auto regressive model and survival analysis methods. The auto regressive model is utilized here to reproduce the advancement of the community structure, while the survival analysis systems permit the expectation of future changes the community may undergo.