Case Studies
A look at Autobox's forecasting solutions
Business Problem: Consider a case where we only have 4 readings with each one taken an hour apart. By using data at each minute we are able to increase our sample size to 240. We are not increasing the number of samples, but the statistical calculation is done as if we have, and so the number of degrees of freedom for the significance test is incorrectly increased and a spurious conclusion is reached. This is one of primary causes of "spurious correlation".  view
Anheuser-Busch
Business Problem: In-Stock conditions not maximized and inventories at retail too large.
The heart of the solution to improving in-stock conditions and minimizing inventories lies in the ability to accurately forecast demand for products at retail. 		view
Presentation: Planning the Final Mile of the Supply Chain at Anheuser-Busch, presented at the IBF conference February 23 & 24, 2004. How Anheuser-Busch partnered with a major retail grocery chain to create orders for individual stores.
 view
Cannibalization Study
Presentation: The process of discovering new causal variables: Is unpredictable. Driven by gleaning patterns from historical forecast errors and outliers identified by Autobox. Often leads to causal variables we didn’t know existed. view
Carreker forecasting solution
Business Problem: Provide systems to banks to estimate required cash at ATM/MAC machines The need for accurate forecasts for Automated Teller Machines is critical so that customer’s can get the cash they need while minimizing the amount of cash at the machine. view
Commerce Bank forecasting solution
Business Problem: Increased losses on credit card charge due to bankruptcy or non payment With the economic slowdown and the unemployment rate, financial institutions are facing new challenges to maintain good financial health. view
Financial analysis/Fraud Detection using Autobox 
International Symposium on Forecasting 2002 presentation on the use of Autobox as a fraud detection mechanism. view
This study evaluates whether automatic intervention detection(AID) can be effectively used to distinguish companies with fraudulent reported data from those with no indication of fraudulent reports. view
General Mills
Business Problem: Develop accurate forecasts and incorporate significant marketing variables (price, TV ads etc.) into a working model that would allow marketing and logistics to more effectively allocate resources. view
Goodyear forecasting solution
Business Problem: Goodyear Engineering Products Division wanted a system that would enable them to automate the inclusion of inputs to accurately determine the “safety stock” for each product to minimize inventory while maintaining promised customer service levels. view
Oak Ridge National Laboratory forecasting solution
Business Problem: The detection of release events in the annual growth increments of trees has become a central and widely applied method for reconstructing the disturbance history of forests. While numerous approaches have been developed for identifying release events, the preponderance of these methods relies on running means that compare the percent change in growth rates. view
Pharmaceutical Company solution
Business Problem: How to determine how many samples to give to a doctor. Sales Reps call on doctor’s to discuss their drug with scientific discussions on why their drug is better. They also leave behind samples for the doctor to distribute to make it easier for them to adopt their product. view
Regression vs Box Jenkins
Case Study: If you are going to analyze time series data perhaps this discussion will be of help. Regression was originally developed for cross-sectional data but Statisticians / Economists have been applying it (mostly incorrectly) to chronological or longitudinal data with little regard for the Gaussian assumptions. view
Switzerland Government
Case Study: The client needs to cleanse their data not forecast. In order to have accurate reporting by Canton (County), they use Autobox to identify pulses and level shifts and the historical data is restated without these effects. view
   
©1999-2009 Automatic Forecasting Systems. All rights reserved.
Terms of use