ABSTRACT
This chapter presents the literature review on flood alert system using high-resolution radar rainfall data. There are many approaches that can be used to predict the future direction and magnitude of a physical process, such as rainfall. Forecasting having two predominant branches: Qualitative Forecasting and Quantitative Forecasting. The autoregressive method is a type of random process, which is used to predict some types of natural phenomena, falling within the group of linear prediction formulas. The moving average method is a way where the current observations depend on all past observations. Mujumdar and Nagesh used two criteria for the model selection Maximum Likelihood (ML) rule and Mean Square Error are used for the selection of the best model for each of the rivers considered. The selection of a model by the ML rule involves evaluating a likelihood value for each of the candidate models and choosing the model which gives the highest value.
