+ Reply to Thread
Results 1 to 8 of 8

Thread: Heteroskedasticity

  1. #1
    Points: 533, Level: 10
    Level completed: 66%, Points required for next Level: 17

    Posts
    50
    Thanks
    20
    Thanked 0 Times in 0 Posts

    Heteroskedasticity




    I found heteroskedasticity in my data and I took the log for my dependent variable in SPSS. My independent variables are dummies).

    To be sure I checked in SPSS again for heteroskedasticity using the Breusch-Pagan Test and the Koenker Test, this is not available in SPSS but I found a Syntax/Matrix.

    The test showed p <.05 Implying I still have heteroskedasticity. I did some more research and found that I can run a regression with Adjusted Standard Errors. But SPSS doesn't have this option.

    I found a syntax, but it doesn't seem to work (maybe because I have some categorical/dummy variables in my model?)

    Any ideas on how to handle heteroscedasticity in SPSS? Maybe I can get access to STATA but that's a BIG maybe

  2. #2
    Omega Contributor
    Points: 38,413, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,004
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Heteroskedasticity

    You could always use R or python, which are free. I find it a little shocking SPSS would not have a robust SE option available. Wait lets see what google finds:


    http://www-01.ibm.com/support/docvie...id=swg21477323


    Well does this work for you?


    P.S., this is my quick search: https://www.google.com/search?q=sand...s_rd=ssl#spf=1
    Stop cowardice, ban guns!

  3. #3
    Points: 1,741, Level: 24
    Level completed: 41%, Points required for next Level: 59

    Posts
    230
    Thanks
    37
    Thanked 68 Times in 59 Posts

    Re: Heteroskedasticity

    Can you do us a favor and post some residual plots? Sometimes, you can have heteroscedasticity that's indicating you need a grouping variable or is caused by one. The former case warrants trying to ID that variable and including it.

  4. #4
    Points: 533, Level: 10
    Level completed: 66%, Points required for next Level: 17

    Posts
    50
    Thanks
    20
    Thanked 0 Times in 0 Posts

    Re: Heteroskedasticity

    Quote Originally Posted by ondansetron View Post
    Can you do us a favor and post some residual plots? Sometimes, you can have heteroscedasticity that's indicating you need a grouping variable or is caused by one. The former case warrants trying to ID that variable and including it.
    I've attached screenshots
    Attached Thumbnails Attached Thumbnails Click image for larger version

Name:	plot4.jpg‎
Views:	17
Size:	14.5 KB
ID:	6545   Click image for larger version

Name:	plot4gr.jpg‎
Views:	9
Size:	8.7 KB
ID:	6546   Click image for larger version

Name:	plot6.jpg‎
Views:	10
Size:	13.9 KB
ID:	6547   Click image for larger version

Name:	plot6gr.jpg‎
Views:	7
Size:	8.7 KB
ID:	6548  

  5. #5
    Points: 1,741, Level: 24
    Level completed: 41%, Points required for next Level: 59

    Posts
    230
    Thanks
    37
    Thanked 68 Times in 59 Posts

    Re: Heteroskedasticity

    Quote Originally Posted by Jazz3 View Post
    I've attached screenshots
    So sorry I forgot to ask if you can use some plotting symbols such as the categorical variables themselves. It kind of looks like the left half and the right half could be two separate groups and at the very least, it's helpful to know what those groups are. What is your dependent variable?

  6. #6
    Points: 533, Level: 10
    Level completed: 66%, Points required for next Level: 17

    Posts
    50
    Thanks
    20
    Thanked 0 Times in 0 Posts

    Re: Heteroskedasticity

    Quote Originally Posted by ondansetron View Post
    So sorry I forgot to ask if you can use some plotting symbols such as the categorical variables themselves. It kind of looks like the left half and the right half could be two separate groups and at the very least, it's helpful to know what those groups are. What is your dependent variable?
    Can I send you a PM?

  7. #7
    Omega Contributor
    Points: 38,413, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,004
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Heteroskedasticity

    What is with the clusters in the scattergraph and the line breaks in the others?
    Stop cowardice, ban guns!

  8. #8
    Points: 533, Level: 10
    Level completed: 66%, Points required for next Level: 17

    Posts
    50
    Thanks
    20
    Thanked 0 Times in 0 Posts

    Re: Heteroskedasticity


    the first two images are with 4 independent dummy variables and some control variables (of which some also dummies) the 3rd and 4th image is with 6 independent dummy variables and the same control variables

+ Reply to 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