Re: Choosing the right method

Eric wrote: 
Dear all,

A frined of mine, who is ophtalmic surgeon needs to make
a study regarding factors affecting a disease.
He has got a dataset of appromimately 300 samples (patients),
with around 6 or 7 explanatory variables (quantitative or qualitative),
and one response variable which is the occuring of the disease.

The problem is that in the dataset the disease is quite rare so that
there are few cases of diseases.

What would be the best method in order to discriminate the populations.

We have been suggested : Artificial Neural Net, K nearest neighbors or Decision tree.

Thank you,

Eric


None of those methods will yield reliable predictions. You'll need something like data reduction (e.g. incomplete principal components logistic regression) or penalized maximum likelihood estimation. Above all don't attempt to find which of the variables are 'significant'

Frank
.

Relevant Pages

  • Fiber Disease
    ... There is a very controversial Disease that is affecting many in our ... communities. ... I am personally affected by this Illness. ...
    (sci.med.pathology)
  • Re: Light for knitting?
    ... disease that is affecting so many of my age-mates. ... remove my glasses to see stuff closer ...
    (rec.crafts.textiles.yarn)
  • Re: Test Results are IN
    ... secondary infections and a myriad of other health problems, affecting ... won't know if the same virus causes Lyme Disease, Lou Gherig"s Disease, ... MS, ALS, Fibromyalga, Chronic Fatigue, Alzheimers, Essential tremors, ...
    (sci.med.diseases.lyme)
  • Re: Towel From Me
    ... The way we are playing I doubt it. ... caught that disease that's affecting the scum at the moment We've got a ...
    (uk.sport.football.clubs.liverpool)
  • Re: Towel From Me
    ... The way we are playing I doubt it. ...
    (uk.sport.football.clubs.liverpool)