Nowadays, the retail industry is very competitive. Most companies in this industry are facing many problems to satisfy customers the most and to be the most efficient. One of the most important problems is to make sales forecasting. In the past, it is more up to experiences to make sales forecasting, therefore the accuracy is often not good. With the development of computer and AI, machine learning methods, in the present, it is easier and more accurate to make a forecast for sales. In this thesis, time series models are applied with the aid of R programming to make sales forecasting. Firstly, we go to understand the basic knowledge about time series models, then we take an example of forecasting sales for a retail shop to apply these methods, including average, naive, snaive, drift, exponential smoothing, ARIMA, dynamic regression models. In the end, we come up with a conclusion about what we did in this thesis.

Brown, R. G. (1959). Statistical forecasting for inventory control. McGraw/Hill.
Galit Shmueli, Kenneth C.Lichtendahl Jr (2015) Practical time series forecasting with R: a hand-on guide. Axelrod Schnall Publishers.
Holt, C. E. (1957). Forecasting seasonals and trends by exponentially weighted averages (O.N.R. Memorandum No. 52). Carnegie Institute of Technology, Pittsburgh USA.
https://www.kaggle.com/c/favorita-grocery-sales-forecasting/overview/description
Hyndman, R.J., & Athanasopoulos, G. (2018) Forecasting: principles and practice, 2nd edition, OTexts: Melbourne, Australia. OTexts.com/fpp2
Wickham, H. (2016). ggplot2: Elegant graphics for data analysis (2nd ed). Springer.
Winters, P. R. (1960). Forecasting sales by exponentially weighted moving averages. Management Science, 6, 324–342.