Ok! Hahaha, R esta molt be :)

Hahaha, si python és el costat fosc, quin és el de la llum?

Thanks @msuzen.bsky.social

Absolutely!! :)

Thanks a lot @alxndrmlk.bsky.social !! Hope you like it!

This is brilliant!

unobserved confounder = ghost node

(I guess it is a reference of music's ghost note, right?)

I also think should be official

You can take a look here livebook.manning.com/book/causal-...

Link to the book: mng.bz/Xxaa

Discount (until December 23rd): pbruizdevilla

github: github.com/aleixrvr/Cau...

short video: shorturl.at/LFCxu

I’m deeply grateful to Manning Publications for giving me the opportunity to write this book. A very special thanks goes to Frances Lefkowitz—she’s been essential in making this book clear and accessible to everyone.

I hope you enjoy it!

There are lots of examples, and I’ve included code in both R and Python, plus an introduction to the #DoubleML library.

The book walks you through Pearl’s graphical approach step by step—how to model reality with graphs and use them to choose the right analysis. It covers the adjustment formula and the backdoor criterion, which are tools to separate correlation from causation.

The second reason is more personal. Are you the kind of person who wants to understand why things are the way they are? If so, CI is for you. Its whole purpose is to uncover the why.

First, making decisions based on data isn’t as easy as it seems. There are always tons of factors affecting things at the same time. We need a way to figure out what really matters and what doesn’t, so we can estimate the impact of our decisions. That’s where CI comes in.

So, why do I think causal inference (CI) is a must-have skill for anyone working with data? Here are my two big reasons.