how to draw a graph of distribution when x axis is continuous varible while y axis is categorical value

[Kai-Lun Hu]

Dear experts
I am new to this forum and I want to thank you in advance for any help.
Varible1 is categorical of yes or no, varible2 is continuous variable with 10000 values. I want to know the estimated rate of yes in each value of variable2 and make the curve smooth. I use "twoway fpfitci varible1 varible2" and it doesn' work.
Thank you
Kai-Lun

[FernandoRios]

Hi Kai
two questions:
1. Why do you think "twoway fpfitci varible1 varible2" does not work?
2. If instead of 10000 variable 2 had only 100 values, how would you calculate what you need to do?
One possible solution. Have you tried "lpoly". Or look into semiparametric regression analysis?
Fernando

[Kai-Lun Hu]

Hi Kai
two questions:
1. Why do you think "twoway fpfitci varible1 varible2" does not work?
2. If instead of 10000 variable 2 had only 100 values, how would you calculate what you need to do?
One possible solution. Have you tried "lpoly". Or look into semiparametric regression analysis?
Fernando

Hi Fernando
Thank you for your response.
Maybe the problem is that I don't know how to change the variable 1 into percentage. logic regression shows that variable 2 is significantly associated with variable1. But I want to use a graph to express.
Kailun

[FernandoRios]

So something like this:

. webuse lbw
two fpfit low age || lpoly low age

which still raises the question, why does lpfit does not "work" for you?
And what about my point 2. How would yo do this, if instead of 10000 values you had only 100, or 20?

[Kai-Lun Hu]

So something like this:

. webuse lbw
two fpfit low age || lpoly low age

which still raises the question, why does lpfit does not "work" for you?
And what about my point 2. How would yo do this, if instead of 10000 values you had only 100, or 20?

If their were only 20, I think it is meaningless to do that. I would rather use variable1 as the x axis to compare the mean value of variable2. in your code, the problem still exists, that is y axis is the exact value of low. what i want is y axis should be the percentage of low birthweight in each value of age.