The K Nearest Neighbor algorithm is an intuitive ML algorithm that is well-suited for nowcasting and forecasting. I have studied how well the algorithm can perform when it comes to producing nowcasts and forecasts of Swedish GDP based in the National Institute of Economic Research's Economic Tendency Survey. The results have been published in articles in Journal of Business Cycle Research, Applied AI Letters, and Journal of Quantitative Economics.
While EViews is a versatile software capable of excellent data handling and econometric modeling, it still lacks some machine learning functionality. However, it is possible to integrate Python-based ML algorithms, such as those in Scikit-learn, directly into EViews.
This document provides a step-by-step guide on setting up such a system, leveraging Python's powerful ML packages to generate nowcasts and forecasts within EViews: EViewsPython.pdf