Title :
Regression Analysis of Intrinsic Linear Models with Automated Transformations of Monotone Predictors
Author :
Tysiak, Wolfgang
Author_Institution :
Appl. Sci. Dortmund Univ., Dortmund
Abstract :
One of the problems in intrinsic linear models is the characterization of the necessary transformations. In this article we show a way to detect several monotone transformations (convex, concave, dummy variables) by means of logistic functions in the predictors.
Keywords :
linear systems; mathematics computing; regression analysis; transfer functions; automated transformations; intrinsic linear models; logistic functions; monotone predictors; regression analysis; Conferences; Data acquisition; Data analysis; Least squares methods; Linear regression; Logistics; Predictive models; Regression analysis; Shape; Tail; data analysis; intrinsic linear models; least squares in the non-linear case; logistic transformation of predictors; regression analysis;
Conference_Titel :
Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, 2005. IDAACS 2005. IEEE
Conference_Location :
Sofia
Print_ISBN :
0-7803-9445-3
Electronic_ISBN :
0-7803-9446-1
DOI :
10.1109/IDAACS.2005.283059
