Best visualization of joint/marginal distributions?

1 reply · 5K views · last reply
nycixc
Messages
1
Reaction score
0
I'm trying to get a better visualization of joint/marginal distributions. It's my weakest conceptual area as I pursue the actuary exams, and I want to fully understand this on a more statistical level.

I've taken linear/diff-eq/multivariate, so I'm completely comfortable with the integration and other menial work involved, but I feel like I am lacking in the area of visualizing the distribution, and therefore am incorrectly identifying the limits needed. I'm used to looking at a shape and identifying the limits that way or being given the limits flat-out.

So for ƒ(x,y), what are the best visualizations you've got for:
ƒx(x), ƒy(y), ƒ(X|Y=y), and ƒ(Y|X=x), given 0 < x < y < 1 or x>0, y>0.(I've seen the marginals represented using Σp(x,y), but anything will help!)
 
https://www.google.com/search?q=Bes...ChMI4de43IydxwIVhnU-Ch2w9go6&biw=1024&bih=653

3D plot of a joint distribution: a scatter of points on the X-Y plane inside a green ellipse, with the marginal density p(Y) in red along the X axis and p(X) in blue along the Y axis, drawn as curves over histogram bars.


3D histogram of a joint distribution over two variables, with red and blue marginal histograms projected onto the back walls and a grey and colored bar grid on the floor, axes labeled -4 to 4 and -6 to 5.