Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am trying to do nonlinear regression on a dataset obtained from full factorial design (2 and 3 levels) with the help of genetic programming.

Is there something fundamentally wrong with trying to construct a nonlinear regression model with full-factorial design? My background is not in statistics, and I did read somewhere that full factorial design (esp when few levels are used) are supposed to be used at most to identify contribution of interaction factors, and key factors, and not for full blown nonlinear regression.

Would appreciate your answers...

share|improve this question
What do you mean under nonlinear regression ... with the help of genetic programming? Could you provide more details? Why don't you simply use polynimial fitting? – Paul Jul 17 '12 at 7:03
genetic programming is able to find a mathematical model of data by the evolutionary computing method of iterative test-generate-select. given a set of operators, and training data (input-output), the program constructs randomly a set of expression trees, refined through many iterations of recombination , mutation, etc, (which in effect is exchange of subtrees), fit evaluation, and fitness based selection to arrive at a model of data. i hope that's somewhat descriptive. i tried regression trees, polynomials, and the result on training and test data is not as good as genetic programming. – oalah Jul 17 '12 at 11:29

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.