A commom technique in time series modeling is to decompose the time series into it s trend and seasonal components. Inside the Score-Driven Models class, this decomposition is usually made in an additive form, so that the series is expressed as the sum of its trend and seasonal components. However, it is not unusual that, even with the seasonal component being considered into the model, the residuals still show signs of seasonal dependency that were not captured by the model. With that said, the main objective of this project is to study if different non-linear combinations of those components are able to improve forecast accuracy in score driven models.