A new method called Reweighted Robust Support Vector Regression (WRSVR) has been developed to estimate regression and approximate functions by gently removing outliers. The approach involves adjusting parameters to improve the regression function iteratively. First, an initial regression function is obtained using Support Vector Regression (SVR), then a weighted SVR objective function is created to find a new regression function. This process is repeated until convergence. The WRSVR method is simple, robust against outliers, and easy to use. Numerical simulations demonstrate that WRSVR outperforms Standard SVR, Robust Support Vector Regression Networks, and Weighted Least Square SVM in terms of robustness.