Tree-structured regression model using a projection pursuit approach

Hyunsun Cho; Eun Kyung Lee

doi:10.3390/app11219885

Tree-structured regression model using a projection pursuit approach

Hyunsun Cho, Eun Kyung Lee

Department of Statistics

Research output: Contribution to journal › Article › peer-review

4 Scopus citations

Abstract

In this paper, a new tree-structured regression model—the projection pursuit regression tree—is proposed. It combines the projection pursuit classification tree with the projection pursuit regression. The main advantage of the projection pursuit regression tree is exploring the independent variable space in each range of the dependent variable. Additionally, it retains the main properties of the projection pursuit classification tree. The projection pursuit regression tree provides several methods of assigning values to the final node, which enhances predictability. It shows better performance than CART in most cases and sometimes beats random forest with a single tree. This development makes it possible to find a better explainable model with reasonable predictability.

Original language	English
Article number	9885
Journal	Applied Sciences (Switzerland)
Volume	11
Issue number	21
DOIs	https://doi.org/10.3390/app11219885
State	Published - 1 Nov 2021

Bibliographical note

Funding Information:
Funding: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (2018R1A2B6001251).

Publisher Copyright:
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.

Keywords

Exploratory data analysis
Piecewise regression
Projection pursuit
Recursive partition
Regression tree

Access to Document

10.3390/app11219885

Cite this

@article{d401f9b0a94c49e892cacf616da3e354,

title = "Tree-structured regression model using a projection pursuit approach",

abstract = "In this paper, a new tree-structured regression model—the projection pursuit regression tree—is proposed. It combines the projection pursuit classification tree with the projection pursuit regression. The main advantage of the projection pursuit regression tree is exploring the independent variable space in each range of the dependent variable. Additionally, it retains the main properties of the projection pursuit classification tree. The projection pursuit regression tree provides several methods of assigning values to the final node, which enhances predictability. It shows better performance than CART in most cases and sometimes beats random forest with a single tree. This development makes it possible to find a better explainable model with reasonable predictability.",

keywords = "Exploratory data analysis, Piecewise regression, Projection pursuit, Recursive partition, Regression tree",

author = "Hyunsun Cho and Lee, {Eun Kyung}",

note = "Funding Information: Funding: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (2018R1A2B6001251). Publisher Copyright: {\textcopyright} 2021 by the authors. Licensee MDPI, Basel, Switzerland.",

year = "2021",

month = nov,

day = "1",

doi = "10.3390/app11219885",

language = "English",

volume = "11",

journal = "Applied Sciences (Switzerland)",

issn = "2076-3417",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "21",

}

TY - JOUR

T1 - Tree-structured regression model using a projection pursuit approach

AU - Cho, Hyunsun

AU - Lee, Eun Kyung

N1 - Funding Information: Funding: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (2018R1A2B6001251). Publisher Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland.

PY - 2021/11/1

Y1 - 2021/11/1

N2 - In this paper, a new tree-structured regression model—the projection pursuit regression tree—is proposed. It combines the projection pursuit classification tree with the projection pursuit regression. The main advantage of the projection pursuit regression tree is exploring the independent variable space in each range of the dependent variable. Additionally, it retains the main properties of the projection pursuit classification tree. The projection pursuit regression tree provides several methods of assigning values to the final node, which enhances predictability. It shows better performance than CART in most cases and sometimes beats random forest with a single tree. This development makes it possible to find a better explainable model with reasonable predictability.

AB - In this paper, a new tree-structured regression model—the projection pursuit regression tree—is proposed. It combines the projection pursuit classification tree with the projection pursuit regression. The main advantage of the projection pursuit regression tree is exploring the independent variable space in each range of the dependent variable. Additionally, it retains the main properties of the projection pursuit classification tree. The projection pursuit regression tree provides several methods of assigning values to the final node, which enhances predictability. It shows better performance than CART in most cases and sometimes beats random forest with a single tree. This development makes it possible to find a better explainable model with reasonable predictability.

KW - Exploratory data analysis

KW - Piecewise regression

KW - Projection pursuit

KW - Recursive partition

KW - Regression tree

UR - http://www.scopus.com/inward/record.url?scp=85117610798&partnerID=8YFLogxK

U2 - 10.3390/app11219885

DO - 10.3390/app11219885

M3 - Article

AN - SCOPUS:85117610798

SN - 2076-3417

VL - 11

JO - Applied Sciences (Switzerland)

JF - Applied Sciences (Switzerland)

IS - 21

M1 - 9885

ER -

Tree-structured regression model using a projection pursuit approach

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this