Luca Bittarello

Negative weights

1

Is there a reason for which we forbid negative weights ([ref](https://github.com/Quantco/glum/blob/8be33850e695bf56e0bb0fafb32e1d3c6fed7249/src/glum/_glm.py#L173))? Some IV weighting schemes use negative weights to isolate compliers.

I find myself importing them from `glum._distribution`, which isn't very sightly.

Intercept-only GLMs

2

It'd be nice to be able to fit GLMs without covariates. It's usually trivial, except when offsets are involved...

Concatenation distorts interactive legend

2

Sorry if this issue has been reported before! Consider this example from [the documentation](https://altair-viz.github.io/user_guide/interactions.html#selection-targets-fields-and-encodings): ```python import altair as alt from vega_datasets import data cars = data.cars.url selection = alt.selection_multi(fields=["Origin"]) color...

[python-package] `verbose` does not supress warnings with custom objectives

11

## Description In the Python package of LightGBM 4.0.0, setting the `verbose` parameter to `-1` does not suppress warnings if the objective is a user-defined function. ## Reproducible example ```python...

This PR restructures the distributions to avoid diamond inheritance -- i.e. the gamma, inverse Gaussian, normal and Poisson distributions no longer inherit from the Tweedie distribution. Because it is often...

Integration with Microeconometrics

2

I'm writing the package [`Microeconometrics`](https://github.com/lbittarello/Microeconometrics.jl), which includes some code to compute covariance matrices. I was wondering if we could merge it into `CovarianceMatrices`, so that I could focus on point...