A new method called Fast Improvised Influential Distance (FIID) has been developed to identify important data points in linear regression models more efficiently. This method helps to better understand and classify different types of outliers in data sets, leading to more accurate results. Additionally, a weighted least squares (WLS) technique based on FIID has been created to handle heteroscedasticity in regression models, improving the estimation process. Overall, these new approaches have been shown to outperform existing methods in terms of accuracy and efficiency.