Retail sales analysis and forecasting

This notebook is an analysis of a retail brand transactions record. The idea was to answer basic questions (How many product were offered on week 50 ? What is the best selling product overall ?) and then try to forecast sales for the upcoming week in the dataset (week 98), using two methods (ARIMA and Random Forest) and comparing them. For further information, do not hesitate to contact me at louis.auneau.c@gmail.com (I am also looking for a data science engineering position !).