Regional economic forecasting based on structural equation modelling and time series

Policymaking, economic planning, and corporate decision-making all benefit from regional economic forecasting. Traditional forecasting techniques, nevertheless, can struggle to handle time dependency and complicated causality and this study thus suggests a regional economic forecasting model based on the combination of structural equation modelling (SEM) and time series analysis (TSA), SEMTSA-Region. The SEMTSA-Region model shows better forecasting performance in several assessment criteria when the model is employed in the experiment to anticipate the economic data of a region in the previous ten years and compared with many conventional models. Furthermore, this work performed parameter optimisation trials on the model to support its great stability and adaptability even more. This paper offers a fresh theoretical framework for regional economic forecasting as well as a useful base for actual implementations.