To minimize the primal support vector machine (SVM) problem, we propose to use iterative majorization. To do so, we propose to use it- erative majorization. To allow for nonlinearity of the predictors, we use (non)monotone spline transformations. An advantage over the usual ker- nel approach in the dual problem is that the variables can be easily inter- preted. We illustrate this with an example from the literature.

, ,
hdl.handle.net/1765/7889
Econometric Institute Research Papers
Report / Econometric Institute, Erasmus University Rotterdam
Erasmus School of Economics

Groenen, P., Bioch, C., & Nalbantov, G. (2006). Nonlinear support vector machines through iterative majorization and I-splines (No. EI 2006-25). Report / Econometric Institute, Erasmus University Rotterdam. Retrieved from http://hdl.handle.net/1765/7889