Forecasting and data mining can become particularly challenging when data:

• Is particularly large or complex
• Has limited history
• Contains a large amount of irrelevant data
• Has pockets of missing information

Advanced forecasting solutions may be more cost-effective for cases where data is particularly challenging. Two examples of advanced forecasting and data mining techniques are:

• Auto_ARIMA: Useful for situations where the data is prone
  to seasonality “spikes” or “outliers”
• Neural Networks: A technique that takes advantage of repetitive
  learning through repeated forecasts and comparisons
  against assumptions. This technique is suited to situations
  where data contains missing values, short time periods,
  inconsistent reporting of data and where the user has multiple
  items to predict