Hugo Delatte
Hugo Delatte
The lint issue has been fixed, the tests should pass the ruff check now
Your syntax is correct. Could you please provide the error message you are receiving? The `GINI_MEAN_DIFFERENCE` risk measure can be slow to solve. Here are some tests I've run with...
I've compared it with Riskfolio; it's the same algorithm, with the only difference being the default parameters. For example, based on the above example and with 300 observations, they both...
Thank you for the suggestion. I agree that we should add more details in the install section. Skfolio uses CVXPY, which has pretty good and straightforward documentation about the installation....
Correct, it should be 252 indeed
I'm working on this
Hello, for a quick fix you can explicitly pass the risk free rate to the predicted portfolio using `portfolio_params`: ```python model = MeanRisk( objective_function=ObjectiveFunction.MAXIMIZE_RATIO, risk_free_rate=0.02 / 252, portfolio_params=dict(risk_free_rate=0.02 / 252)...
Thank you! Likewise, both `sktime` and `skpro` are amazing. That's definitely something we need to explore. I'll start working on examples that use `sktime` estimators to see how it fits...
Thank you for the resources! Regarding our current design approach, I'll provide some examples and list the pros and cons I've identified. Let's consider the following example, where we build...
@fkiraly thank you for the detailed design thoughts. As you mentioned, we may need to relax type checking when using polymorphic estimators. I'll work on concrete use cases to find...