Let's begin by looking at traditional time series forecasting techniques, specifically ARIMA and exponential smoothing to model demand in simple use cases. We will look at how ARIMA estimates sales using historical sales and forecast errors. Also, we'll review how exponential smoothing accounts for irregularities in historical sales and captures trends and seasonality to forecast sales.
Understanding traditional time series forecasting
Auto-Regressive Integrated Moving Average (ARIMA )
ARIMA is a time series analytical technique used to capture different temporal structures in univariate data. To model the time series data, differencing is applied across the series to make the data stationary. Differencing is the...
