An Algorithm for the forward step of adaptive regression slines via mapping approach

Download
2012
Kartal Koç, Elçin
In high dimensional data modeling, Multivariate Adaptive Regression Splines (MARS) is a well-known nonparametric regression technique to approximate the nonlinear relationship between a response variable and the predictors with the help of splines. MARS uses piecewise linear basis functions which are separated from each other with breaking points (knots) for function estimation. The model estimating function is generated in two stepwise procedures: forward selection and backward elimination. In the first step, a general model including too many basis functions so the knot points are generated; and in the second one, the least contributing basis functions to the overall fit are eliminated. In the conventional adaptive spline procedure, knots are selected from a set of distinct data points that makes the forward selection procedure computationally expensive and leads to high local variance. To avoid these drawbacks, it is possible to select the knot points from a subset of data points, which leads to data reduction. In this study, a new method (called S-FMARS) is proposed to select the knot points by using a self organizing map-based approach which transforms the original data points to a lower dimensional space. Thus, less number of knot points is enabled to be evaluated for model building in the forward selection of MARS algorithm. The results obtained from simulated datasets and of six real-world datasets show that the proposed method is time efficient in model construction without degrading the model accuracy and prediction performance. In this study, the proposed approach is implemented to MARS and CMARS methods as an alternative to their forward step to improve them by decreasing their computing time

Suggestions

Restructuring forward step of MARS algorithm using a new knot selection procedure based on a mapping approach
Koc, Elcin Kartal; İyigün, Cem (2014-09-01)
In high dimensional data modeling, Multivariate Adaptive Regression Splines (MARS) is a popular nonparametric regression technique used to define the nonlinear relationship between a response variable and the predictors with the help of splines. MARS uses piecewise linear functions for local fit and apply an adaptive procedure to select the number and location of breaking points (called knots). The function estimation is basically generated via a two-stepwise procedure: forward selection and backward elimin...
An analysis of a linearly extrapolated BDF2 subgrid artificial viscosity method for incompressible flows
Demir, Medine (Elsevier BV, 2020-10-01)
This report extends the mathematical support of a subgrid artificial viscosity (SAV) method to simulate the incompressible Navier-Stokes equations to better performing a linearly extrapolated BDF2 (BDF2LE) time discretization. The method considers the viscous term as a combination of the vorticity and the grad-div stabilization term. SAV method introduces global stabilization by adding a term, then anti-diffuses through the extra mixed variables. We present a detailed analysis of conservation laws, includin...
EVALUATING THE CMARS PERFORMANCE FOR MODELING NONLINEARITIES
Batmaz, İnci; Kartal-Koc, Elcin; Köksal, Gülser (2010-02-04)
Multivariate Adaptive Regression Splines (MARS) is a very popular nonparametric regression method particularly useful for modeling nonlinear relationships that may exist among the variables. Recently, we developed CMARS method as an alternative to backward stepwise part of the MARS algorithm. Comparative studies have indicated that CMARS performs better than MARS for modeling nonlinear relationships. In those studies, however, only main and two-factor interaction effects were sufficient to model the nonline...
Refinements, extensions and modern applications of conic multivariate adaptive regression splines
Yerlikaya Özkurt, Fatma; Weber, Gerhard Wilhelm; Department of Scientific Computing (2013)
Conic Multivariate Adaptive Regression Splines (CMARS) which has been developed at the Institute of Applied Mathematics, METU, as an alternative approach to the well-known data mining tool Multivariate Adaptive Regression Splines (MARS). CMARS is based on given data and a penalized residual sum of squares for MARS, interpreted as a Tikhonov Regularization problem. CMARS treats this problem by a continuous optimization technique called Conic Quadratic Programming (CQP). This doctoral thesis adapts the CMARS ...
An improved method for inference of piecewise linear systems by detecting jumps using derivative estimation
Selcuk, A. M.; Öktem, Hüseyin Avni (Elsevier BV, 2009-08-01)
Inference of dynamical systems using piecewise linear models is a promising active research area. Most of the investigations in this field have been stimulated by the research in functional genomics. In this article we study the inference problem in piecewise linear systems. We propose first identifying the state transitions by detecting the jumps of the derivative estimates, then finding the guard conditions of the state transitions (thresholds) from the values of the state variables at the state transitio...
Citation Formats
E. Kartal Koç, “An Algorithm for the forward step of adaptive regression slines via mapping approach,” Ph.D. - Doctoral Program, Middle East Technical University, 2012.