@nelkhadiri Profile picture

نوار الخديري

@nelkhadiri

when being Human doesn't work, try being Alien !

Joined December 2010
Similar User
Mohamed Amine BERGACH photo

@AminemS

Issam OURRAI photo

@Samourrai

Onzyme photo

@onzyme

hamid khalifa photo

@mido_twitt

Dr. H photo

@DocteurH

RRC photo

@RRC_Terra

Adam photo

@iMaroc

نوار الخديري Reposted

[Thread] Comment signaler ce genre de lâche sur la nouvelle plateforme de la police marocaine : E-Blagh.ma ?


نوار الخديري Reposted

#Maroc 🇲🇦 - En plein #Casablanca, Tout s’est arrêté pour sauver ce petit être vivant … Vive notre peuple ! 🥰🕊️


نوار الخديري Reposted

“Nonreplicable” publications are cited more than “replicable” ones? statmodeling.stat.columbia.edu/2024/06/02/non…


نوار الخديري Reposted

You get what you measure

Tweet Image 1

نوار الخديري Reposted

To all DIDstas: is your Y measured less frequently than your D? E.g. you look at US election outcomes, measured every 4 years, but treatment measured every year? Check out great vignette @diegociccia1 has prepared on how to use did_multiplegt_dyn then: github.com/chaisemartinPa…


نوار الخديري Reposted

I come back to this a lot

Tweet Image 1

نوار الخديري Reposted

MCMC draws cannot fill the posterior in high dimensions statmodeling.stat.columbia.edu/2024/05/06/mcm…


نوار الخديري Reposted

“You want to gather data to determine which of two students is a better basketball shooter. You plan to have each student take N shots. Roughly how large does N have to be for you to have a good chance of distinguishing a 30% shooter from a 40% shooter?”statmodeling.stat.columbia.edu/2024/05/04/you/


نوار الخديري Reposted

Hi #EconTwitter! Always nice to see an #econometrics paper forthcoming in Econometrica, especially if it's about the bootstrap! This cool work by Ayden Higgins (@UniofOxford) & Koen Jochmans (@TSEinfo), building on earlier research by @silviag45129862 & Kaffo (2016,…

Tweet Image 1

نوار الخديري Reposted

You have to be a leading lady but you can’t be nominated for best actress. You have to be a strong female director but you can’t be nominated for best director. You have to have a wildly popular dance song but you can’t be nominated for best original song.

Tweet Image 1

نوار الخديري Reposted

Statistical inference in information geometry: Convert data into empirical distribution, and project it onto model submanifold wrt divergence: - Kullback-Leibler divergence projection: Maximum Likelihood Estimator - reverse KLD: Maximum Entropy 👉PDF: ams.org//journals/noti…

Tweet Image 1

نوار الخديري Reposted

To support my students and anyone learning #DataScience, I built out an interactive #Python dashboard with @matplotlib to demonstrate Monte Carlo methods. We use random sampling as a general solution for difficult or expensive deterministic problems! In this case we solve a…


نوار الخديري Reposted

Two masterpiece books on INFORMATION THEORY worth your reading time! 🎉 MacKay book PDF is available online with video lectures! 👉 inference.org.uk/itila/book.html

Tweet Image 1

نوار الخديري Reposted

Marginal, joint & conditional, probabilities & distributions. Fundamental @statistics concepts and the building blocks of #DataAnalytics & #MachineLearning. I built out a set of #Python @matplotlib interactive #dashboards to help you understand these fundamental concepts on…


نوار الخديري Reposted

Hi #EconTwitter! In '83, @Prof_RobEngle, Henrdy & Richard published this ground-breaking #econometrics paper on 𝐄𝐱𝐨𝐠𝐞𝐧𝐞𝐢𝐭𝐲 in econometric models. How does it correlate with current exogeneity concepts in mainstream econometrics & applied economics? A must-read!

Tweet Image 1

نوار الخديري Reposted

Hi #EconTwitter! Interested in 𝐜𝐚𝐮𝐬𝐚𝐥 𝐢𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞? Check out this brand new JRSSb #Statistics paper, where the focus is in causal effects on distributions, rather than summary measures such as ATE. New concept: the average causal effect map! Cool stuff 😀

Tweet Image 1

نوار الخديري Reposted

Our team proposes a new causal definition of agency that allows us to discover whether an agent is present in a system. This leads to better causal modelling of AI agents and their incentives, improving the safety analysis of a system before training it: dpmd.ai/discovering-ag…

Tweet Image 1

نوار الخديري Reposted

🚩The false discovery rate (FDR) is designed to quantify the trade-off between true positives and false positives. A λ can be chosen to try to balance this trade-off, making it particularly useful for performing high-dimensional hypothesis tests.(Storey) 🔗theta.edu.pl/wp-content/upl…


United States Trends
Loading...

Something went wrong.


Something went wrong.