Genalg: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Benjamin
No edit summary
imported>Benjamin
No edit summary
 
Line 29: Line 29:
===See Also===
===See Also===


[[selectvars]], [[calibsel]], [[fullsearch]], [[gaselctr]], [[genalgplot]], [[ipls]], [[rpls]], [[sratio]], [[vip]], [[Genetic Algorithms for Variable Selection]]
[[selectvars]], [[calibsel]], [[fullsearch]], [[gaselctr]], [[genalgplot]], [[ipls]], [[rpls]], [[sratio]], [[vip]], [[Genetic Algorithms for Variable Selection]], [[Sample and Variable Selection]], [[Variable Selection]]

Latest revision as of 10:44, 11 January 2018

Purpose

Genetic algorithm for variable selection to optimize model predictive ability with graphical user interface.

Synopsis

genalg(xdat,ydat)

Description

GENALG performs variable selection using a genetic algorithm. The function creates a graphical user interface that allows the user to load data from the workspace and select all of the GA algorihtm optional parameters (GASELCTR is a command-line version). A wide range of GA settings can be selected from the GUI. Please see gaselctr for a description of each option or Genetic Algorithms for Variable Selection for a discussion of GA use.

Optional inputs are the training data consisting of a matrix of predictor variables xdat and either a column vector or a matrix of predicted variable ydat. (The number of rows in xdat and ydat must be the same). If GENALG is called with no inputs, xdat and ydat can be loaded using the File menu.

In addition to various plots, the GUI can produce and save the results in a model structure that is the same as that returned by GASELCTR. Please see GASELCTR for a description of the model. Also, if "settings" are saved from GENALG this is the same as the options structure discussed in GASELCTR.

Optional Inputs

  • xdat = x-block
  • ydat = y-block

Examples

  >> x2 = mncn(x);
  >> y2 = mncn(y);
  >> genalg(x2,y2)

See Also

selectvars, calibsel, fullsearch, gaselctr, genalgplot, ipls, rpls, sratio, vip, Genetic Algorithms for Variable Selection, Sample and Variable Selection, Variable Selection