Dynamic Causal Effects in Econometrics with a Focus on the Nonparametric Method: A Review Paper

According to recent developments and research in causal inference applications, understanding applied modeling in causal effects is particularly important in econometrics. we also provide an outline of econometrics' use of causal inference. while most economists recognize randomized controlled experiments as the preferred method for concluding, a considerable portion of empirical research in econometrics relies on observational data, which presents challenges such as confounding and potential loss of exogeneity. in this context, we examine two contemporary research types: randomized experiments and observational studies. when linear estimators are biased towards dynamic causal effects due to carryover facts or serial correlation in the imputation mechanism, applying a nonparametric framework is suggested. because the nonparametric framework does not rely on functional assumptions about the underlying data-generating processes, such as linearity or limited. our review of the dynamic causality study approach, the linear method, which includes lp and var, and nonlinear statistical modeling, which provides for bart and their use in econometrics, are all reviewed in this paper. dynamic systems modeled using linear parametric models often encounter limitations impacting forecasting accuracy and policy implications. bart specifications can produce more precise tail forecasts than the var structure on the nonparametric framework. finally, bart has the lowest rmse in linear and nonlinear data generation processes, and the performance of bart essential variables in a macroeconomic data set is optimal compared to other regression estimators.