<p>This paper aims to propose an M-estimation approach for estimating parameters of a nonlinear censored regression model, specifically the Tobit model, while addressing the issue of endogeneity. We employ a two-stage procedure to deal with endogeneity. In the first stage, the control function approach is utilized, incorporating instrumental variables to mitigate endogeneity. The second stage involves M-estimation for estimating the unknown parameters of the model. We derive strong consistency for the proposed estimator. Additionally, the finite sample performance of the estimators is evaluated through Monte Carlo simulations. The analysis is also conducted on two distinct datasets to implement the proposed methodology, allowing for a robust evaluation of the model’s performance across different scenarios. To evaluate model performance, the bootstrap method is employed to compute Mean Squared Errors (MSE) for parameter estimates.</p>