In the area of machine learning, most of the algorithms are intended for small n large p problems. I am familiar with the statistical techniques of PCA, etc but was wondering what algorithms are available in the machine learning area for similar analysis work. The aim is to determine the effect of a limited number of predictor variables on the outcome using robust methods such as pca in the ML domain,

Thanks in advance.


closed as unclear what you're asking by Scortchi, Xi'an, chl Feb 6 '15 at 9:27

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 1
    $\begingroup$ p referring to ? $\endgroup$ – Peter Ellis Jul 6 '13 at 3:45
  • 1
    $\begingroup$ It may be helpful to search this site for: lasso, elastic net, support vector machines; and perhaps supervised PCA and partial least squares in relation to dimension reduction followed by prediction. All of these handle the case where p (number variables) is larger than n (number samples). $\endgroup$ – julieth Jul 6 '13 at 4:01
  • 1
    $\begingroup$ @julieth - the OP is asking for large n small p (n>>p), not the othe way around. $\endgroup$ – probabilityislogic Jul 6 '13 at 4:54
  • 1
    $\begingroup$ I voted to close as a duplicate, but this was based on a mis-reading the title as asking about $n<p$ and not about $n>p$. Now I see that it is not a duplicate, but instead an unclear question. If I could, I would vote to close as unclear. $\endgroup$ – amoeba Jan 26 '15 at 15:03
  • 1
    $\begingroup$ A large sample size & few potential predictors is the relatively unproblematic case. So wouldn't the answer just be a list of supervised learning methods? $\endgroup$ – Scortchi Jan 30 '15 at 9:49