+ Reply to Thread
Results 1 to 3 of 3

Thread: robust regression estimators

  1. #1
    Points: 279, Level: 5
    Level completed: 58%, Points required for next Level: 21

    Location
    south borneo
    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    robust regression estimators




    Hi everyone.

    I have studied the robust regression a few weeks ago and I want to have a lot of insight about it. After reading some sources, I know some estimators on robust regression, i.e. M-estimators, MM-estimators, LTS-estimators, LMS-estimators, and s-estimators. However, every source I read just a little explanation about this. My question is "What distinguishes it from theory and application on the estimators?"
    What is the usefulness of robust regression in addition to detecting outliers ?

    thanks

  2. #2
    Omega Contributor
    Points: 36,707, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Activity Award
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,828
    Thanks
    392
    Thanked 1,158 Times in 1,119 Posts

    Re: robust regression estimators

    Well, I don't think it is so much the detection of outliers as the ability not to let outliers influence the results.


    I need to better learn about robust regression, I would appreciate it if you could list some of the better articles or websites links. Thanks.
    Stop cowardice, ban guns!

  3. #3
    Fortran must die
    Points: 55,021, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,514
    Thanks
    688
    Thanked 915 Times in 874 Posts

    Re: robust regression estimators


    One thing to remember about robust regression is that it only works with interval DV.

    This is a SAS paper on robust regression http://www2.sas.com/proceedings/sugi27/p265-27.pdf

    This is a UCLA site that discusses robust regression for SAS http://www.ats.ucla.edu/stat/sas/dae/rreg.htm

    A discussion of the background http://web.as.uky.edu/statistics/use...notes/12-1.pdf

    A discussion of procrobustreg which is the sas product that deals with this http://www.phusewiki.org/docs/2006/ST01.pdf

    I stopped working on this when it became obvious most of my DV were categorical in nature.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  4. The Following User Says Thank You to noetsi For This Useful Post:

    hlsmith (01-30-2017)

+ 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