Application of Linear Fuzzy and Ordinary Linear Regression on the Geographical data with Outlier Observation Case Study: Saghez Station

In regression models normally, both of data and parameters are considered as crisp. But, in some cases, there is obscurity in the model parameter or observations. In these cases fuzzy regression can be fair alternative model. In this research we applied the mentioned models to forecasting Temperature (response variable) of Saghez areas. To do this we consider the Wet Temperature (WT), Relative Humidity (RH) and Cloud Angle (CA) as descriptive variables. Finally the estimated models and the parameters show the high determination coefficient and significance values to forecast the temperature. apply these approaches to geography data (WT, RH, WS, CA) with symmetric triangular fuzzy response observations.

Keywords:

Linear Fuzzy Regression , Absolute Value and least square Regression , Climatology , Precipitation