if the conditions are met, the Ordinary Least Squares (OLS) approach in a linear regression model is regarded the best method for estimating the linear regression parameters. On the other hand, the findings will be misleading if the data does not match the underlying assumptions. The presence of outliers and heteroscedastic errors in the data causes the assumption of constant variance in the least squares regression to be violated. In linear regression, where the least squares estimators satisfy the property of minimum variance, this assumption of constant variance is applied (homoscedasticity) is critical. To cope with the problem of outliers in the data, a robust regression method is necessary when the assumption of error variance is violated. In this paper will utilize Weighted Least Square (WLS) estimation to estimate the parameter of regression coefficients. In a transformed variables technique, WLS estimation is the same as OLS estimation. Outliers can have a big impact on the WLS to handle this; we proposed a robust technique for estimating regression parameters when heteroscedastic and outliers are present. To estimate the model parameters, we utilize the robust regression estimator of least trimmed squares and robust regression of M-estimation utilizing Huber and Tukey's Iterative Reweighted Least Squares (IRWLS). In 1993, the model parameters of Altitudes toward the biggest problems facing public school status, United States, 1990-93a were estimated. According to the findings, estimators obtained from the least trimmed method and M-estimation techniques are more successful than OLS estimators