+ Reply to Thread
Results 1 to 6 of 6

Thread: please help : modelization

  1. #1
    Points: 19, Level: 1
    Level completed: 37%, Points required for next Level: 31

    Posts
    4
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Exclamation please help : modelization




    Hello,

    I have a distribution which looks like a Khi-2 distribution, which model is most appropriate?

    I try linear regression, but normal distribution is not suitable.

    Thanks for your help !

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: please help : modelization

    I may be wrong, but isn't the chi-sq = to N * N, so can you just examine the square root of your outcome variable?
    Stop cowardice, ban guns!

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

    elle1301 (07-22-2016)

  4. #3
    Points: 19, Level: 1
    Level completed: 37%, Points required for next Level: 31

    Posts
    4
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: please help : modelization

    oh it's a good idea ! I will try and then I write here if it worked
    thank you

  5. #4
    Points: 19, Level: 1
    Level completed: 37%, Points required for next Level: 31

    Posts
    4
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: please help : modelization

    So, the distribution more looks like a normal distribution, but the error of linear model increase, I will continue to search but if you have other idea I'll take it.

  6. #5
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: please help : modelization

    Well it doesn't matter too much if the y-variable itself is normally distributed. You just need the assumptions about the model error term to hold. So you should be working to normalize them and get homoscedasticity!
    Stop cowardice, ban guns!

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

    elle1301 (07-22-2016)

  8. #6
    Points: 19, Level: 1
    Level completed: 37%, Points required for next Level: 31

    Posts
    4
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: please help : modelization


    Oh yes it's true, so I will continue to search new explanatory variables

+ 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