Abstract: Multicollinearity remains a significant concern in Simultaneous Equation Models, as the inherent interdependence of variables within the system can lead to high levels of correlation among the independent variables. This can have substantial implications for the reliability and interpretation of parameter estimates in the Simultaneous Equation Models. As such, this paper propose an extended two stage least squares (2sls) estimator by introducing principal components which will eliminate the concern of high correlation among the variables in the simultaneous equation model. The extended 2sls estimator (2sls-pc estimator) transforms the predictors to principal components before producing an estimate. In a bid to compare the classical two stage least squares (2sls) with the two stage least squares with principal components (2sls-pc), simultaneous equation model with three equations predicting Final Consumption Expenditure, Gross Domestic Investment and Gross Domestic Product were modelled. The 2sls-pc estimator addressed the high collinearity between the dataset and produced more significant estimate than the classical 2sls estimator while retaining all the information from the original dataset. Keywords: Endogeneity, Endogenous, Exogenous, Multicollinearity, Principal Components, SEM, Two stage Least square. Title: A Robust Estimator for Causal Inference: Integrating Two Stage Least Squares with Principal Component Author: Toba Temitope Bamidele, Alabi Olatayo Olusegun International Journal of Recent Research in Mathematics Computer Science and Information Technology ISSN 2350-1022 Vol. 11, Issue 1, April 2024 - September 2024 Page No: 27-32 Paper Publications Website: www.paperpublications.org Published Date: 06-July-2024 DOI: https://doi.org/10.5281/zenodo.12671069 Paper Download Link (Source) https://www.paperpublications.org/upload/book/A%20Robust%20Estimator%20for%20Casual-06072024-2.pdf