An attempt has been made to model global solar radiation using the multiple linear regression method with various meteorological parameters as input. The meteorological parameters used in this study are the relative humidity, wind speed, maximum, minimum and mean temperatures, sunshine hours, and the clearness index. The global solar radiation on a horizontal surface measured at three different locations representing the varying climatic conditions prevailing over TamilNadu has been analyzed. The study sites are Chennai, Kodaikanal and Coimbatore. In our study, first we developed a multiple regression model to estimate the global radiation at each site separately. Then, we developed an overall multiple regression equation for the estimation of global radiation anywhere in TamilNadu. The robust least square regression technique has been used in our study. This method is preferred as it is vulnerable to the outliers in the data set. The performance of the regression models was evaluated using the following statistical error parameters: Mean Absolute Percentage Error (MAPE), Mean Bias Error (MBE), Root Mean Squared Error (RMSE) and T-Statics(TS).