Model Selection with Stepwise Regression
While no model selection criteria is perfect, making use of the avaliable statistical tools allows a researcher to systematically choose a set of predictive variables for a given set of data and critiria. The selection process can be done by an autmoatic procedure in the form of a sequence of tests such as F-tests or making use of the Akaike information criterion.
"The most that can be expected from any model is that it can supply a useful approximation to reality: All models are wrong; some models are useful". -- George Box
Requirements
- R: environment for statistical computing and graphics
- car: Companion to Applied Regression
- gplot: Various R Programming Tools for Plotting Data
References
- D. W. Higinbotham et al., Physical Review C93 (2016) 055207.
- R Core Team, R: A Language and Environment for Statistical Computing.
- J. Fox and S. Weisberg, An R Companion to Applied Regression.
BibTeX
@article{Higinbotham:2015rja,
author = "Higinbotham, Douglas W. and Kabir, Al Amin and Lin,
Vincent and Meekins, David and Norum, Blaine and Sawatzky,
Brad",
title = "{Proton Radius from Electron Scattering Data}",
journal = "Phys. Rev.",
volume = "C93",
year = "2016",
number = "5",
pages = "055207",
doi = "10.1103/PhysRevC.93.055207",
eprint = "1510.01293",
archivePrefix = "arXiv",
primaryClass = "nucl-ex",
reportNumber = "JLAB-PHY-16-2",
SLACcitation = "%%CITATION = ARXIV:1510.01293;%%"
}
@Manual{R,
title = {R: A Language and Environment for Statistical Computing},
author = {{R Core Team}},
organization = {R Foundation for Statistical Computing},
address = {Vienna, Austria},
year = {2013},
note = {{ISBN} 3-900051-07-0},
url = {http://www.R-project.org/},
}
@Book{car,
title = {An {R} Companion to Applied Regression},
edition = {Second},
author = {John Fox and Sanford Weisberg},
year = {2011},
publisher = {Sage},
address = {Thousand Oaks {CA}},
url = {http://socserv.socsci.mcmaster.ca/jfox/Books/Companion},
}