search

maximenc / pycop

Python library for multivariate dependence modeling with Copulas
https://pypi.org/project/pycop/
MIT License
97 stars 20 forks source link

_ #12

Closed ghost closed 7 months ago

maximenc commented 1 year ago

Hi there,

After estimating your parameter(s), the CDF can be obtained using the copula function (Sklar’s theorem):

$$ H(x,y) = C \left( F_X(x), F_Y(y) \right) $$

To access the CDF, use: cop.get_cdf(u, v, param) (u, v in [0, 1])

For the LLF, things get trickier. While the CMLE doesn't account for the marginals, you would need the PDF of the marginals.

The LLF can then be derived from the joint PDF:

$$ h(x,y) = c \left( F_X(x), F_Y(y) \right) f_X(x) f_Y(y) $$

The copula PDF can be accessed with: cop.get_pdf(u, v, param)

I hope this helps.

M.

jondo commented 8 months ago

It seems this issue can be closed.