Neural networks are worth surveying as part of the extended data mining and modeling toolkit. Of particular interest is the comparison of more traditional tools like regression analysis to neural networks as applied to empirical model-building.
A cable company uses logistic regression to determine the variables most predictive of a "truck roll" (technician visit to customer's home) within seven days of a new installation.
quick and easy-to-remember way for Lean Six Sigma practitioners to get the most benefit from simple linear regression analysis is with a simple check-up method. The method borrows and adapts the familiar concept found in the 5S tool.
Applying regression requires special attention from the analyst. Each process step needs to be carefully examined and executed; a small mistake may lead to an erroneous model. This article describes some common mistakes made in regression and their corresponding remedies.
Linear regression is often considered difficult or confusing by those practitioners just beginning to delve into the Six Sigma toolkit. Making sense of it starts at a basic level.
Sometimes Six Sigma practitioners find a Y that is discrete and Xs that are continuous. How then can a regression equation be developed? The correct technique is something called logistic regression, but this tool is often not well understood.
An organization had a limited capital budget but needed to ensure its product was safely delivered in its own packaging. The organization used poka-yoke and realistic tolerancing to control its process improvements.
While Black Belts often make use of R-Squared in regression models, many ignore or are unaware of its function in ANOVA models or GLMs. Input variables may then be overvalued, which may not lead to a significant improvement in the Y.