Open Access
Regression model selection using genetic algorithms
Sandra Paterlini,Tommaso Minerva +1 more
- 13 Jun 2010
- pp 19-27
TL;DR: GARST turns out to be even better compared to GARS as variable transformations help to improve results further, and the type of transformations revealed the relationships between dependent and independent variables.
read more
Abstract: The selection of independent variables in a regression model is often a challenging problem. Ideally, one would like to obtain the most adequate regression model. This task can be tackled with techniques such as expert based selection, stepwise regression and stochastic search heuristics, such as genetic algorithms (GA). In this study, we investigate the performance of two GAs for regressors selection (GARS) and for regressors selection with transformation of the regressors (GARST). We compare the performance with stepwise regression for the "Fat Measurement" and the "Cholesterol Measurement" datasets and use the AIC, BIC and SIC statistical criteria to quantify the adequacy of the models. The results for GARS are superior for all statistical criteria compared to both forward and backward stepwise regression, but not always when R2 and RMSE statistics are considered. GARST turns out to be even better compared to GARS as variable transformations help to improve results further. Moreover, the type of transformations revealed the relationships between dependent and independent variables.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
A survey on evolutionary machine learning
Harith Al-Sahaf,Ying Bi,Qi Chen,Andrew Lensen,Yi Mei,Yanan Sun,Binh Q. Tran,Bing Xue,Mengjie Zhang +8 more
TL;DR: This paper provides a review on evolutionary machine learning techniques for major machine learning tasks such as classification, regression and clustering, and emerging topics including combinatorial optimisation, computer vision, deep learning, transfer learning, and ensemble learning.
Tax payment default prediction using genetic algorithm-based variable selection
TL;DR: Results show that variables measuring solvency, liquidity and payment period of trade payables are important variables in predicting tax defaults, and the best performing model comprises three non-linearly transformed variables and has a predictive accuracy of 73.8%.
41
Variable selection in Logistic regression model with genetic algorithm.
Zhongheng Zhang,Victor Trevino,Sayed Shahabuddin Hoseini,Smaranda Belciug,Arumugam Manivanna Boopathi,Ping Zhang,Florin Gorunescu,Florin Gorunescu,Velappan Subha,Songshi Dai +9 more
TL;DR: This tutorial paper aims to provide a step-by-step approach to the use of Genetic algorithms (GA) in variable selection, which can be extended and adapted to other data analysis needs.
Methods for variable selection in LiDAR-assisted forest inventories
Paolo Moser,Alexander Christian Vibrans,Ronald E. McRoberts,Erik Næsset,Terje Gobakken,Gherardo Chirici,Matteo Mura,Marco Marchetti +7 more
TL;DR: In this paper, the authors compared three methods to select predictor variables for nonlinear models of relationships between forest attributes and LiDAR metrics, two of them based on genetic algorithms (GAs) and one based on random forest (RM).
34
Privacy Preserving Linear Regression on Distributed Databases
TL;DR: A (theoretical) privacy preserving linear regression model for the analysis of data owned by several sources is presented and uses a semi-trusted third party and delivers on privacy and complexity.
27
References
Building ARMA Models with Genetic Algorithms
Tommaso Minerva,Irene Poli +1 more
- 18 Apr 2001
TL;DR: A genetic algorithm is built which evolves the representation of a predictive model, choosing both the orders and the predictors of the model.
33
A genetic algorithm for graphical model selection
Irene Poli,Alberto Roverato +1 more
TL;DR: This paper implements the search procedure as a genetic algorithm and proposes a crossover operator which operates on subgraphs and shows it to perform better than an automatic backward elimination procedure at the cost of a small increase of computational time.
16
Evolutionary approaches for statistical modelling
Tommaso Minerva,Sandra Paterlini +1 more
- 12 May 2002
TL;DR: This paper proposes an approach to select multivariate linear regression models as well as to build ARMA time-series models and introduces a methodology to tackle the clustering problem in a model-based framework.
11
Technological modelling for graphical models: an approach based on genetic algorithms
TL;DR: An automatic model search procedure for the identification of an optimal set of good models is proposed and it is shown that the identified models can co-exist, whereas in a scientific modelling approach such models represent a starting point for further context-dependent analysis.
10
Regression Analysis: Theory, Methods, and Applications.
Abstract: An up-to-date, rigorous, and lucid treatment of the theory, methods, and applications of regression analysis, and thus ideally suited for those interested in the theory as well as those whose interests lie primarily with applications. It is further enhanced through real-life examples drawn from many disciplines, showing the difficulties typically encountered in the practice of regression analysis. Consequently, this book provides a sound foundation in the theory of this important subject.