+ Reply to Thread
Results 1 to 6 of 6

Thread: logistic regression : intercept is NOT significative

  1. #1
    Points: 2,896, Level: 32
    Level completed: 98%, Points required for next Level: 4

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    logistic regression : intercept is NOT significative




    Hi everybody

    I have performed a logistic regression with a stepwise selection. But, at the end, the parameter estimate of intercept is NOT significative.

    What should I do ?

    Thanks for the help

  2. #2
    TS Contributor
    Points: 21,765, Level: 92
    Level completed: 42%, Points required for next Level: 585

    Location
    Israel
    Posts
    393
    Thanks
    0
    Thanked 7 Times in 7 Posts
    I am not sure there is much you can do, usually it's not advisable to remove the intercept, even when it's not significantly different than 0, since it can make your prediction biased. did you try other model selection methods rather than stepwise ?

  3. #3
    TS Contributor
    Points: 8,362, Level: 61
    Level completed: 71%, Points required for next Level: 88

    Location
    Crete, Greece
    Posts
    717
    Thanks
    0
    Thanked 35 Times in 34 Posts
    Weeg is right, do not remove the constant.

  4. #4
    Points: 2,521, Level: 30
    Level completed: 48%, Points required for next Level: 79

    Posts
    20
    Thanks
    0
    Thanked 0 Times in 0 Posts
    You do not need to worry that the intercept beta0 hat is insignificant. What you need to consider seriously is the regression coefficients of each predictor variables. They must all pass the significance test. As to the intercept term, just leave it.

  5. #5
    R purist
    Points: 35,103, Level: 100
    Level completed: 0%, Points required for next Level: 0
    TheEcologist's Avatar
    Location
    United States
    Posts
    1,921
    Thanks
    303
    Thanked 607 Times in 341 Posts
    Quote Originally Posted by a little boy View Post
    You do not need to worry that the intercept beta0 hat is insignificant. What you need to consider seriously is the regression coefficients of each predictor variables. They must all pass the significance test. As to the intercept term, just leave it.
    Depends on what you want to know.

    In some cases people are actually interested in the intercept. e.g. base performance in absence of variable x (x=0) or height of different species of offspring at birth ect ect....

    In that case it certainly matters, but in 90% in the cases it isn't important as 'a little boy' states; people are mostly interested linear relationship between variables. So you need to ask yourself what do you want to know and how does the intercept fit in.
    The true ideals of great philosophies always seem to get lost somewhere along the road..

  6. #6
    Points: 2,521, Level: 30
    Level completed: 48%, Points required for next Level: 79

    Posts
    20
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Quote Originally Posted by TheEcologist View Post
    Depends on what you want to know.

    In some cases people are actually interested in the intercept. e.g. base performance in absence of variable x (x=0) or height of different species of offspring at birth ect ect....

    In that case it certainly matters, but in 90% in the cases it isn't important as 'a little boy' states; people are mostly interested linear relationship between variables. So you need to ask yourself what do you want to know and how does the intercept fit in.
    Thank you TheEcologist, I learned a lot!

+ Reply to Thread

           




Similar Threads

  1. Replies: 3
    Last Post: 04-13-2010, 08:29 AM
  2. Intercept in Regression
    By zarasandreas in forum Statistics
    Replies: 1
    Last Post: 01-09-2010, 03:29 PM
  3. Replies: 3
    Last Post: 10-23-2009, 11:52 AM
  4. Linear regression without intercept
    By Douglas Sousa in forum Statistics
    Replies: 0
    Last Post: 05-13-2009, 01:27 PM
  5. Regression: When the intercept is zero.
    By Michelle in forum Statistics
    Replies: 0
    Last Post: 06-24-2006, 08:25 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