Bivariate copula#3
Open
alvaror9554 wants to merge 19 commits into
Open
Conversation
- Fixed hardcoded 200 parameter bounds to use Frank-specific 20 bounds - Enhanced Gram matrix regularization with adaptive conditioning - Added weight validation safeguards for numerical stability - Removed scaler-related NaN sources through proper bounds handling - Frank copula now produces clean coefficients without NaN values
Author
|
Changes in online_mvdistreg |
- Removed obsolete test files and data - Frank copula working with stable coefficients - All NaN issues resolved with proper parameter bounds
alvaror9554
force-pushed
the
bivariate_copula
branch
from
e37cc87 to
adac321
Compare
- Replace slow list comprehensions with vectorized st.t.ppf calls - Vectorize st.t.pdf calculations for major speedup - Apply optimizations to likelihood and derivative functions - Achieve 4.1x performance improvement (90s -> 22s) - Maintain 100% mathematical accuracy with same analytical formulas
- Change rho initialization from direct tau to sin(tau * pi/2) transformation - Change nu default from 10 to 8 to match R VineCopula MLE method - Improves compatibility with R BiCopEst initialization values - Addresses numerical stability and convergence consistency
- Integrated R VineCopula compatibility improvements in bicop_studentt.py - Applied performance optimizations from remote branch - Resolved conflicts while preserving improved initial_values method
- Removed leftover merge conflict marker that was causing syntax errors - File now compiles cleanly without red error indicators
…into bivariate_copula
… compatible implementation
… DF iteration work
…ion from chrischu12
…ion, and performance improvements
… into bivariate_copula
chrischu12
reviewed
Feb 17, 2026
|
|
||
| # Calculate the gamma ratio using stable division | ||
| gamma_ratio = nu/2 | ||
| gamma_ratio = stable_gamma_division((nu + 2.0) / 2.0, nu / 2.0) |
Author
There was a problem hiding this comment.
The change for nu has been corrected
chrischu12
reviewed
Feb 17, 2026
| except: | ||
| continue | ||
|
|
||
| def negative_log_likelihood(params): |
Owner
There was a problem hiding this comment.
you define this in many functions at the same time, please define only once in the t copula
chrischu12
reviewed
Feb 17, 2026
| def get_regularization_size(self, dim: int) -> int: | ||
| return dim | ||
|
|
||
| def r_style_nu_optimization(self, y, initial_rho=None): |
Author
There was a problem hiding this comment.
This was a function to test the initial values in R, it is of no use now, I have already deleted it.
chrischu12
reviewed
Feb 17, 2026
| out = prediction | ||
| return out | ||
|
|
||
| def _df_iteration(self, y, rho_values): |
Owner
There was a problem hiding this comment.
this must go in the copula function no?
Author
There was a problem hiding this comment.
Yes it should go into the copula. I already moved into bicop_studentt.py
chrischu12
reviewed
Feb 17, 2026
| rho_updated = theta[a][0].flatten() | ||
|
|
||
| # Optimize nu using DF iteration | ||
| optimal_nu = self._df_iteration(y, rho_updated) |
Owner
There was a problem hiding this comment.
but this means you also reoptimize even when you estimate nu using covariates, no?
Author
There was a problem hiding this comment.
Yes, now we also re-optimize nu when covariates are used, so the nu update is refreshed with new data in this setup
chrischu12
force-pushed
the
bivariate_copula
branch
from
f958ffd to
b5207eb
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Make a coment in the second derivative function.