Sales Forecast for Amazon Sales with Time Series Modeling

Accurate sales prediction plays an important role in reducing costs and improving customer service levels, especially for B2C(Business to consumer) e-commerce. This paper attempts to forecast future sales at Amazon.com, Inc. based on historical sales data. Firstly, it proposes three possible forecasting approaches according to the historical data pattern, that is Holt-Winters exponential smoothing, neural network auto regression model and ARIMA(Autoregressive integrated moving average). Secondly, it specifies certain accuracy measures using which well determine the suitability of the forecast methods on the available sales data. Finally the three methods will be implemented to forecast Amazons quarterly sales in 2019. The results can help Amazon well manage its future operations.