Nonsmooth DC programming approach to clusterwise linear regression: optimality conditions and algorithms

The clusterwise linear regression problem is formulated as a nonsmooth nonconvex optimization problem using the squared regression error function. The objective function in this problem is represented as a difference of convex functions. Optimality conditions are derived, and an algorithm is designed based on such a representation. An incremental approach is proposed to generate starting solutions. The algorithm is tested on small to large data sets.

[1]  V. F. Demʹi︠a︡nov,et al.  Constructive nonsmooth analysis , 1995 .

[2]  M. Wedel,et al.  Consumer benefit segmentation using clusterwise linear regression , 1989 .

[3]  Le Thi Hoai An,et al.  The DC (Difference of Convex Functions) Programming and DCA Revisited with DC Models of Real World Nonconvex Optimization Problems , 2005, Ann. Oper. Res..

[4]  Pierre Hansen,et al.  Globally optimal clusterwise regression by mixed logical-quadratic programming , 2011, Eur. J. Oper. Res..

[5]  Adil M. Bagirov,et al.  Nonsmooth Optimization Algorithm for Solving Clusterwise Linear Regression Problems , 2015, J. Optim. Theory Appl..

[6]  Adil M. Bagirov,et al.  Nonsmooth DC programming approach to the minimum sum-of-squares clustering problems , 2016, Pattern Recognit..

[7]  Adil M. Bagirov,et al.  An algorithm for clusterwise linear regression based on smoothing techniques , 2015, Optim. Lett..

[8]  Pierre Hansen,et al.  Extensions to the repetitive branch and bound algorithm for globally optimal clusterwise regression , 2011, Comput. Oper. Res..

[9]  R. Horst,et al.  DC Programming: Overview , 1999 .

[10]  Gilbert Saporta,et al.  Clusterwise PLS regression on a stochastic process , 2002, Comput. Stat. Data Anal..

[11]  Luis Angel García-Escudero,et al.  Computational Statistics and Data Analysis Robust Clusterwise Linear Regression through Trimming , 2022 .

[12]  T. P. Dinh,et al.  Convex analysis approach to d.c. programming: Theory, Algorithm and Applications , 1997 .

[13]  Wayne S. DeSarbo,et al.  A simulated annealing methodology for clusterwise linear regression , 1989 .

[14]  Adil M. Bagirov,et al.  Introduction to Nonsmooth Optimization , 2014 .

[15]  F. Clarke.  Optimization And Nonsmooth Analysis , 1983 .

[16]  Padhraic Smyth,et al.  Trajectory clustering with mixtures of regression models , 1999, KDD '99.

[17]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[18]  A. Bagirov,et al.  Discrete Gradient Method: Derivative-Free Method for Nonsmooth Optimization , 2008 .

[19]  Adil M. Bagirov,et al.  Nonsmooth nonconvex optimization approach to clusterwise linear regression problems , 2013, Eur. J. Oper. Res..

[20]  R. Rockafellar.  Convex Analysis: (pms-28) , 1970 .

[21]  丸山 徹.  Convex Analysisの二,三の進展について , 1977 .

[22]  W. DeSarbo,et al.  A maximum likelihood methodology for clusterwise linear regression , 1988 .

[23]  Sven Strauss,et al.  Convex Analysis And Global Optimization , 2016 .

[24]  Helmuth Späth,et al.  Algorithm 39 Clusterwise linear regression , 1979, Computing.