Evaluating Regression Models for Project Forecasting: A Comparative Analysis of Linear and Polynomial Approaches with Monte Carlo Simulations

In this study, we investigate regression prediction schemes and apply them to planning as an extension to predicting future outcomes of projects using Monte Carlo simulations. Project completion probabilities were modeled using cumulative distribution functions (CDF) with linear and polynomial regression models. Linear regression was conducted first, but residuals revealed a high prediction error and the focus was quickly shifted to polynomial regression which increased accuracy substantially. Polynomial regression outperformed the others in capturing the non-linear data patterns and demonstrated low values of mean squared error (MSE) and root mean squared error (RMSE) by making the fitted values close to the actual prices of the test set. Offering trustworthy estimates on project completion this way helps improve decision-making. Thus, further exploration of non-linear methods might aid in maximizing the accuracy of the model.