+ Reply to Thread
Results 1 to 4 of 4

Thread: How do I assess which is the most suitable model?

  1. #1
    Points: 25, Level: 1
    Level completed: 49%, Points required for next Level: 25

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Post How do I assess which is the most suitable model?




    Hey everyone,

    I am making an R project, and I have computed the models using all the methods we've studied in the lectures (particularly Lasso and Forward stepwise selection).
    In the conclusion, I have to assess which is the most suitable specification model.

    Could you help me writing about 10 lines, emphasizing which are the reasons that brought to me to choose one model over the others?

    Thank you.

  2. #2
    Omega Contributor
    Points: 38,432, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,006
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: How do I assess which is the most suitable model?

    Sounds fun. Did you all use validation methods as well (e.g., bootstrap, validation data spits, k-fold CV, LOOCV)? Did you cover parsimony and minimizing error?
    Stop cowardice, ban guns!

  3. #3
    Points: 25, Level: 1
    Level completed: 49%, Points required for next Level: 25

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: How do I assess which is the most suitable model?

    No, I had to compute Lasso, Ridge, stepwise selection and best subset selection on R, without going deeper.
    I've done it, now I have to write 10 lines about how do I select one model over the others, underlining the reasons that brought me to make that decision.

    It's a statistics course for beginners. Also, English is not my mother language so that's why I'm asking for help in writing that.

  4. #4
    Omega Contributor
    Points: 38,432, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,006
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: How do I assess which is the most suitable model?


    It all depends on what you covered in classes for benefits. Typically lasso is preferred if there were close to as many predictors as observations. Though in most applications they all preform the same. You can look at the model's accuracy, but that is not always the best in that the most accurate model me be overfitted and not generalizable. It is hard to provide more information than that without seeing the results.
    Stop cowardice, ban guns!

+ Reply to Thread

           




Tags for this 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