+ Reply to Thread
Results 1 to 2 of 2

Thread: Model to use, fractions as independent and dependent variables

  1. #1
    Points: 5, Level: 1
    Level completed: 9%, Points required for next Level: 45

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Post Model to use, fractions as independent and dependent variables




    Hi,

    I have some background on doing regression analysis, however never on fractional variables. My question is related to model selection on fractional data.

    I have collected a data set that consists of fractions, for example I am trying to predict the amount of fixed assets on balance sheet (independent variable). This figure is calculated and represented as % on total balance sheet value.

    The dependent variables include, but are not limited to, fractions as well, such as the amount of debt (again represented as % of total BS value). Would this model be adversely influenced if continuous non fractional dependents are introduced in the model? Such as company/industry specific dummy variable.

    Obviously plugging these in to a linear regression model does not work, how should I then proceed? I know that in Stata 14 fractional regression is doable, however I only have access to Stata 13 and R.

    I would prefer regression analysis, but if this is too challenging on my data set what other models/analyses would you recommend?

    Thanks,
    ddali

  2. #2
    Points: 3,631, Level: 37
    Level completed: 88%, Points required for next Level: 19
    staassis's Avatar
    Location
    New York
    Posts
    226
    Thanks
    2
    Thanked 41 Times in 39 Posts

    Re: Model to use, fractions as independent and dependent variables


    This is the case for a generalized linear model. You can try beta-distribution as the distribution for the dependent variable. You shpuld be guided by goodness-of-fit limitations of the candidate models and not by programmatic limitations of Stata. I would do it in R or Matlab, where any statistical idea can be expressed relatively easily.

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

    GretaGarbo (11-14-2015)

+ 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