Fashion sales forecasting experiments

To evaluate the performanee of the proposed forecasting model, extensive experiments were conducted in terms of real fashion sales data, which forecast the total sales amounts of various item categories and cities on a monthly, quarterly or annual basis. 

Exterrsive experiments were condircted to validate the proposed HI model in terms of real fashion retail data The experimental resirlts have shown that the HI model can tackle the medium-term sales forecasting problem effectively, which also demorrstrates that the proposed model can provide much superior performance over traditional ARIM A models and two recently developed sales forecasting NN models. A further experiment was presented based on seven irregular annual data sets from M3 competitiort, which further validates the effectiveness of the proposed HI model and shows that the HI model is more powerful to tackle the time series with sufficient sample data. Firrthermore, since the time series tackled in this chapter involve various patterns such as irregirlarity and seasonality, the proposed model is widely applicable and can be easily extended to solve other forecasting problems with similar time-series patterns. 

---