hliebert

It would be great if you could support `felm` from `library(lfe)`. `lfe` supports linear regression with high-dimensional fixed effects, instrumental variables regression and clustering. Having `ivreg` support, as mentioned above,...

Here's another vote for putting this on CRAN. My work mirrors CRAN and access to other remote repositories is blocked. Putting it on CRAN would be very helpful.

> @hliebert I remember the conversation (issue) was `shrtcts` dynamically change its package content in order to store and execute shortcut commands, but CRAN explicitly disallow it in the CRAN...

Got it, thanks for clarifying. Thank you for your work! Whenever you get around to submitting to CRAN, your effort is greatly appreciated.

I am also facing this problem, using the settings from pip.ini would be great.

Upon further reflection, I've updated the first comment to be a bit clearer. I think this also only occurring in combination with cross validation. In this simple example, the recursive...

> The direct forecaster will predict 1 directly, 2 directly, and so on. > > The recursive forecaster will predict 1, then recursively 1 and 2, then recursively 1, 2,...

> First of all, we should probably exclude that the differences are due to algorithm inherent randomness, but I assume you've already ran some experiments to do this? I had...

(Simple example to show the second issue is probably just `random_state`) ```python import numpy as np from sklearn.linear_model import Ridge from sktime.datasets import load_longley from sktime.forecasting.base import ForecastingHorizon from sktime.forecasting.compose...

Since reporting this, I've had a closer look at what the reduction forecasters do under the hood and now believe this is just a symptom of the bugs that are...