Chapter 8. Microsoft Time Series Algorithm

Suppose you are running a retail store and are managing the inventory of each of the products you stock. You know that at the end of the week, people buy more wine, so you put in an extra order and tell your supplier to send the same amount each week. Approaching Valentine's Day, you know that there will be some extra demand, so you order even more for that period. A new, promising winery has only been on the shelves a few weeks, but you see how other wineries from the same region are doing, so you guesstimate how much of that brand to order as well. Also, as wine sales increase, so do sales of cheeses, breads, and gourmet foods, so you increase those orders by a similar amount.

In each of these cases, you are doing time series analysis. You are using the past sales history of product sales to predict future inventory needs. You automatically apply seasonality, including weekly periods and annual events—you even adapt behavior of existing product sales to predict the volume of new sales.

The ability to accurately forecast time series is essential to running almost any business. The Microsoft Time Series algorithm provides a unique approach to time series forecasting that is both intuitive and accurate. This chapter covers the following topics:

  • An overview of the Microsoft Time Series algorithm

  • The usage and application of the Time Series algorithm

  • DMX syntax for time series scenarios

  • Details on how the algorithm works and its parameters

To assist ...

Get Data Mining with Microsoft® SQL Server® 2008 now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.