Tree regression models using statistical testing and mixed integer programming
نویسندگان
چکیده
Abstract Regression analysis is a statistical procedure that fits mathematical function to set of data in order capture the relationship between dependent and independent variables. In tree regression, structures are constructed by repeated splits input space into two subsets, creating if-then-else rules. Such models popular literature due their ability be computed quickly simple interpretations. This work introduces regression algorithm exploits an optimisation model existing method called Mathematical Programming Tree ( MPtree ) optimally split nodes subsets applies test assess quality partitioning. Additionally, approach splitting using multivariate decision rules explored this compared terms performance computational efficiency. Finally, novel introduced performs subset selection on each node select optimal variables considered for splitting, improves proposed algorithm.
منابع مشابه
Parameter Estimation and Change Detection in Linear Regression Models Using Mixed Integer Linear Programming
We present a method for change detection and parameter estimation in change in the mean models and AR(X) model. The method is based on the assumption of piecewise constant parameters resulting in a sparse structure of their derivative. To illustrate the algorithm and the performance of it, we apply it to the change in the mean model and compare it with four other change detection algorithms. Tw...
متن کاملOptimization of Timed Automata Models Using Mixed-Integer Programming
Research on optimization of timed systems, as e.g. for computing optimal schedules of manufacturing processes, has lead to approaches that mainly fall into the following two categories: On one side, mixed integer programming (MIP) techniques have been developed to successfully solve scheduling problems of moderate to medium size. On the other side, reachability algorithms extended by the evalua...
متن کاملAlgorithms for Stochastic Mixed-Integer Programming Models
In this chapter, we will study algorithms for both two-stage as well as multi-stage stochastic mixed-integer programs. We present stagewise (resourcedirective) decomposition methods for two-stage models, and scenario (pricedirective) decomposition methods for multi-stage models. The manner in which these models are decomposed relies not only on the specific data elements that are random, but al...
متن کاملpath planning of manipulator robots using mixed integer nonlinear programming
in this paper a new method is proposed for path planning of planar manipulators amid obstacles through mathematical programming in a way that the robot’s links avoid collision with obstacles throughout their motion from an initial to a goal configuration. after inputting the workspace geometry, the shortest feasible path for the robot’s end-effector is planned toward its goal position using gen...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Computers & Industrial Engineering
سال: 2021
ISSN: ['0360-8352', '1879-0550']
DOI: https://doi.org/10.1016/j.cie.2020.107059