A robust method of forecasting based on fuzzy time series

Present study proposes an improved and versatile method of forecasting based on the concept fuzzy time series forecasting. The developed model has been presented in a form of simple computational algorithms. It utilizes various difference parameters being implemented on current state for forecasting the next state values to accommodate the possible vagueness in the data in a better way and making it a robust method. The developed model has been implemented on the historical student enrollments data of University of Alabama (adapted by Song and Chissom) and the obtained forecasted values have been compared with the existing methods to show its superiority. The robustness of the model has also been tested in comparison. The suitability of the developed model has also been examined in the crop production forecasting by implementing it on historical time series data of rice production of Pantnagar(Farm), India.