+ Reply to Thread
Results 1 to 2 of 2

Thread: Find the Most Important Variables In Predictive Models

  1. #1
    Points: 22, Level: 1
    Level completed: 43%, Points required for next Level: 28

    Thanked 0 Times in 0 Posts

    Find the Most Important Variables In Predictive Models


    I have a target variable Y, and twenty predictors (input variables) X1,X2,...,X20. I can predict Y by its predictors X1,X2,...,X20. Predictors are either categorical or continuous. I want to pick 5 top predictors which have the most effect on predicting the target variable.

    Which statistical method and R-package you recommend to pick the most affecting variables in this prediction?

    With many thanks,

    P.S. SPSS modeler always shows predictor importance when you choose a model such as neural net. However, in R and it's algorithm is my question.

  2. #2
    Omega Contributor
    Points: 38,374, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Not Ames, IA
    Thanked 1,186 Times in 1,147 Posts

    Re: Find the Most Important Variables In Predictive Models

    Many people frown on stepwise regression, if you already know these are significant predictors and are independent of each other, you may want to just use it in this case. Don't know an R package or procedure for this.
    Stop cowardice, ban guns!

  3. The Following User Says Thank You to hlsmith For This Useful Post:

    mojassameh (06-09-2015)

+ Reply to Thread


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Advertise on Talk Stats