+ Reply to Thread
Results 1 to 5 of 5

Thread: Logistic regression analysis and quantitative discrete variables

  1. #1
    Points: 60, Level: 1
    Level completed: 20%, Points required for next Level: 40

    Location
    Presidente Prudente, São Paulo, Brasil
    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Logistic regression analysis and quantitative discrete variables




    Maybe I should start this post explaining that I am a physician working in the public health system in Brazil with little knowledge about the subject underlying my question. I usually follow the tutorials Epi-Info 07 (www.cdc.gov/epiinfo) as help to solve problems of logistic regression. These tutorials work very well when dealing with a binary outcome and other dependent variables that are also binary. However, the task I am trying to solve now involves logistic regression that must include a binary outcome variable and a mix of other qualitative and quantitative discrete dependent variables. Thus, I wonder if exists some kind of logistic regression analysis technique that uses a binary outcome variable while handling both types of dependent variables, quantitative discrete as well as categorical or binary? Thanks for your attention.

  2. #2
    Devorador de queso
    Points: 95,819, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,935
    Thanks
    307
    Thanked 2,629 Times in 2,245 Posts

    Re: Logistic regression analysis and quantitative discrete variables

    I think terminology might be messed up here. "Dependent variable" is usually used to refer to the 'outcome' or the value we're trying to predict. We use "independent variable" to refer to the variables that will be used to model the outcome. Were you using "dependent" to refer to "independent"?
    I don't have emotions and sometimes that makes me very sad.

  3. #3
    Points: 206, Level: 4
    Level completed: 12%, Points required for next Level: 44

    Posts
    23
    Thanks
    6
    Thanked 0 Times in 0 Posts

    Re: Logistic regression analysis and quantitative discrete variables

    I assume you meant independent and not dependent variables. A generalized linear model can handle a mix of independent variable types (categorical, continuous, binary, etc).

  4. #4
    Points: 2,109, Level: 27
    Level completed: 73%, Points required for next Level: 41
    ted00's Avatar
    Location
    USA
    Posts
    237
    Thanks
    21
    Thanked 29 Times in 25 Posts

    Re: Logistic regression analysis and quantitative discrete variables

    in general this can absolutely be done.
    Seems EpiInfo is capable: https://wwwn.cdc.gov/epiinfo/user-gu...C-Command.html
    R is free; g00gle for "r logistic regression" returned several nice examples of all sorts of logistic regression in R

  5. #5
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Logistic regression analysis and quantitative discrete variables


    I am not sure I understand the question. Logistic regression can handle nominal (what I think you mean by qualitative) dependent variables with more than two levels. This is multinominal logistic regression. It can also handle ordinal dependent variables (these might also be interval I assume although I have not seen that addressed). If your variable is interval and has enough levels (different possible values) you would commonly use linear regression although how many unique levels of the DV you need to use linear regression appears to be something statisticians are unsure of or disagree on. Beyond a certain number of levels of the DV analysis of logistic regression becomes cumbersome.

    Also ordinal logistic regression assumes that it does not matter how you would dichotomize the DV - there is a test for this assumption. If it is not met you should use multinominal logistic regression even if your data is ordered.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

+ 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