This chapter explores statistical estimation within regression models. We introduce a comprehensive class of estimators known as Minimum Divergence Estimators (MDEs), along with their empirical loss functions under a parametric framework. Standard properties such as unbiasedness, consistency, and asymptotic normality of these estimators are thoroughly examined. Additionally, the chapter addresses the issue of model misspecification, which can result in biases, inaccurate inferences, and diminished statistical power, and highlights the vulnerability of conventional methods to such misspecifications. Our primary goal is to identify estimators that remain robust against potential biases arising from model misspecification. We place particular emphasis on the \(\gamma \) -divergence, which underpins the \(\gamma \) -estimator known for its efficiency and robustness.