The article presents a new way to predict inflation in Tunisia using a model that considers a lot of different information. This model, called Dynamic Factors Model, gives more accurate predictions of inflation compared to older models in the short term. The researchers split inflation into two types, "free inflation" and "administered inflation", and tested the model on both. They found that the Dynamic Factors Model significantly improves forecasting compared to simpler methods.