Symbolic regression for better specification

Created by W.Langdon from gp-bibliography.bib Revision:1.8129

@Article{TSIONAS:2020:IJHM,

• author = "Mike G. Tsionas and A. George Assaf",
• title = "Symbolic regression for better specification",
• journal = "International Journal of Hospitality Management",
• volume = "91",
• pages = "102638",
• year = "2020",
• ISSN = "0278-4319",
• DOI = "doi:10.1016/j.ijhm.2020.102638",
• URL = "http://www.sciencedirect.com/science/article/pii/S0278431920301900",
• keywords = "genetic algorithms, genetic programming, Symbolic Regression, Specification",
• abstract = "This note introduces the concept of symbolic regression (SR) to tourism and hospitality research. SR uses genetic programming to find the model that best fits the data without a need to pre-specify a functional form or to impose a certain model as a starting point. In other words, SR helps to uncover the intrinsic characteristics of the data at hand. Our view is that SR can serve as an improved method of testing for misspecification. In this note, we propose to derive the true functional form of the residual using SR. We then use this information to improve the forecasts of the linear regression model and, to perform hypothesis tests if needed",

}

Genetic Programming entries for Mike G Tsionas A George Assaf

Citations