+ Reply to Thread
Results 1 to 4 of 4

Thread: Regression Assumptions for Binary Dependent Variable Models?

  1. #1
    Points: 1,209, Level: 19
    Level completed: 9%, Points required for next Level: 91

    Posts
    13
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Regression Assumptions for Binary Dependent Variable Models?




    I was wondering if anyone could direct me to a good source on this, or just answer my question directly.

    What are the assumptions made for a probit or logit model? I'm familiar with assumptions for linear regression, but it's obvious that most of them don't apply to binary dependent variable models (you won't expect a linear X-Y relationship, homoskedasticity, a normally distributed error term, etc.) The only unambigous assumption seems to be that you will have independent observations.

    Is this correct, or am I missing something?

    TIA,

    DB

  2. #2
    Dark Knight
    Points: 6,762, Level: 54
    Level completed: 6%, Points required for next Level: 188
    vinux's Avatar
    Posts
    2,011
    Thanks
    52
    Thanked 241 Times in 205 Posts

    Re: Regression Assumptions for Binary Dependent Variable Models?

    In logistic regression, the distribution assumption(on error) is not required. also homoskedasticity assumption. But other assumptions are still apply... like multicollinearity, linear relationship with logit/probit transformation.. etc

    Following link may add more information
    http://www.statisticssolutions.com/r...tic-regression
    In the long run, we're all dead.

  3. #3
    Points: 1,102, Level: 18
    Level completed: 2%, Points required for next Level: 98

    Location
    near Boston
    Posts
    20
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Regression Assumptions for Binary Dependent Variable Models?

    While Richie's suggestion has some good points, I have found
    http://faculty.chass.ncsu.edu/garson...tic.htm#assume
    to be deeper, more complete, and a little more accurate.

  4. #4
    Points: 1,209, Level: 19
    Level completed: 9%, Points required for next Level: 91

    Posts
    13
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Regression Assumptions for Binary Dependent Variable Models?


    Excellent--your help is much appreciated.

+ Reply to Thread

           




Similar Threads

  1. Regression with an Index as a dependent variable
    By LillyLawson in forum Regression Analysis
    Replies: 2
    Last Post: 03-31-2014, 06:13 AM
  2. Replies: 0
    Last Post: 04-02-2010, 08:13 PM
  3. Replies: 1
    Last Post: 10-22-2009, 11:51 AM
  4. Replies: 7
    Last Post: 05-02-2009, 11:31 AM
  5. Convert censored survival data to binary dependent variable.
    By Davellll in forum Psychology Statistics
    Replies: 1
    Last Post: 06-26-2007, 01:27 PM

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