Assessing the Predictive Accuracy with Machine Learning and Structural Relationship of FDI in Bangladesh Dr. Sayed Mohibul Hossen1, Kapashia Binte Giash1, Md. Mizanoor Rahman1* 1Department of Statistics, Mawlana Bhashani Science and Technology University, Tangail-1902, Bangladesh. Presestating Author: Kapashia Binte Giash Corresponding Auther: Md. Mizanoor Rahman *Corresponding Author’s email: st19009@mbstu.ac.bd Abstract Foreign Direct Investment (FDI) plays a significant role in the economic landscape of a nation. For that reason, policymakers and investigators require to predict FDI and understand the structure of the relationship among variables before making any decision. In recent years, several techniques have been used to predict FDI but limited research has found the structural relationship. This study aims to evaluate the predictive accuracy as well as the structural relationship among FDI and other financial variables. This study analyzed secondary data from the World Bank for the period 1976–2023, which included several variables such as FDI, GDP, Exchange, Trade, GNI, Reserve, Invest, and so on. To predict FDI, we used machine learning techniques including multiple linear regression (MLR), polynomial regression (PR), and structural equation modeling (SEM). The model’s performance was assessed by AIC, BIC, MSE, R-squared, and RMSE with R programming. Our empirical findings revealed that there is a dynamic, non-linear relationship between FDI and other financial variables. This non-linear relationship can be easily handled by structural equation modeling (SEM). It is observed that structural equation modeling (SEM) performs better than the other models and gives a better structural relationship between FDI and other financial variables. To assist policymakers and investigators, it is suggested to apply structural equation modeling (SEM) for investigating structural relationships and predicting FDI in Bangladesh before making any decision. Keywords: Structural Equation Modeling (SEM); Structural Relationship; FDI; Machine Learning (ML);