A Review on Sales on Large-scale Retailing Types in South Korea

In this work, we investigate several time series models to forecast of sales at department stores and at discount stores in South Korea. Considering two large-scale retailing types are homogeneous and comparable in size based on 86 realizations taken from January in 2010 to February in 2017. The dataset of this work is got from Korean Statistical Information Service (KOSIS) and Survey of Service Trend of Korea Statistical Administration, and we exploit the procedures of IBM SPSS 23.0 such as TSPLOT, TSMODEL, EXSMOOTH, ARIMA and MODELFIT. Several model-fit measures, simultaneously, are also used as criteria of the optimal model construction. Time series models can not only predict future sales but can account for economic or business mechanism and increase our understanding of relationships among phenomena or variables under consideration.