Results 1 to 2 of 2

Thread: Checking assumptions regression categorical predictors

  1. #1
    Newbie
    Joined
    Nov 2015
    From
    Netherlands
    Posts
    7

    Checking assumptions regression categorical predictors

    Hi,
    I am doing a regression analysis on review star rating of VR headsets as a dependent variable and 14 attribute dummies. Each dummy is an attribute such as price, resolution, comfort with 3-1 levels (positive mentioned in review, negative mentioned in review, not mentioned). I am now checking assumptions for the linear regression. I have read on this forum that the linearity can't possibly be violated but I can't get it to work for the assumptions homoscedasticity and normally distrubuted errors. I have read some other posts about doing some test but I havent succeed in applying them correctly in SPSS myself. Can anyone help me in checking these assumptions for my research?
    Thanks!
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Newbie
    Joined
    Jul 2017
    From
    New York
    Posts
    5

    Re: Checking assumptions regression categorical predictors

    In your data set the residuals are likely to be non-normal and heteroskedastic. Diagnostic tests will only confirm that... You should focus on getting as much data as possible. If the normality and homoscedasticity assumptions are violated but the sample size is big, the coefficient estimates are accurate and the corresponding t-tests / F-tests are valid. This is due to the Central Limit Theorem and consistency of the Ordinary Least Squares (OLS) procedure.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Replies: 0
    Last Post: Feb 6th 2014, 04:09 AM
  2. OLS assumptions
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: Jul 4th 2012, 11:05 AM
  3. Assumptions of Set Theory
    Posted in the Discrete Math Forum
    Replies: 17
    Last Post: Nov 27th 2011, 06:42 PM
  4. Linear predictors
    Posted in the Advanced Statistics Forum
    Replies: 6
    Last Post: May 28th 2011, 07:11 AM
  5. Categorical and binary responses in multiple linear regression
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: Feb 16th 2011, 01:59 PM

/mathhelpforum @mathhelpforum