0

I am working on a paper demonstrating the use of logistic regression. I am trying to demonstrate that a regular linear probability model is insufficient when the dependent variable is dichotomous as the error terms doesn't represent a normal distribution.

I got two variables in my dataset:

  • gender: Male, Female
  • height: A continuous variable containing the height of every observation.

My logistic model is trying to predict whether the observation is male or female based on their height.

I guess the distribution would be some kind of double gaussian distribution but how do I plot this visually?

SnupSnurre
  • 363
  • 2
  • 12
  • 2
    It's easier to help you if you include a simple [reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) with sample input and desired output that can be used to test and verify possible solutions. – MrFlick Apr 19 '20 at 19:47
  • Perhaps this can help: https://stackoverflow.com/questions/27009641/plot-multiple-normal-curves-in-same-plot – MrFlick Apr 19 '20 at 19:48
  • Complicating further, males and females are inherently differently distributed as to height as well as other relative bodily dimensions as demonstrated in geometric morphometric studies, see especially [Julien Claude](https://www.academia.edu/23070501/Height_and_body_mass_influence_on_human_body_outlines_A_quantitative_approach_using_an_elliptic_Fourier_analysis) perhaps for a cite. – Chris Apr 20 '20 at 03:15

0 Answers0