Statistical tools for forecasting purpose started using smooth exponential methods in 1950s. These methods were modified depending upon the trend followed in the data sets, based upon the evaluation purpose. From simple additive to multiplicative effects and then automated functions were used to evaluate the complexity in data for forecasting purpose. In this review we summarized the various statistical methods used for forecasting purposes starting from the basic function to complex function in order to evaluate various data sets viz-a-viz time series data of different components, like agricultural products, business outcomes, and stock market exchange rates. In order to evaluate the data sets for forecasting purpose to accuracy or near accuracy, various statistical methods will give different predictions depending up on the range of data sets whether daily, weekly, monthly or yearly, number of observations in the data set, seasonality in data sets, number of missing observation in data sets, and more importantly the variation in data sets to interpret the results.