Ernest #Nuamah, des stats qui parlent
#Stata
(O&L : https://olplus.fr/LeIx6)

Taking the small victories today: tamed #stata putexcel command with matrices then used pybliometrics to simplify a #scopus #api search. Being a #data consultant is a little of everything.

Calling #chatgpt from inside of #Stata https://blog.stata.com/2023/07/25/a-stata-command-to-run-chatgpt/ @stata

What are y’all’s favorite academic/research related newsletters or blogs?

#education #academia #stats #psychology #research #rstats #stata

Gov dataset documentation sometimes says that estimates with relative standard errors greater than 30% are unreliable and/or supressed. I'm trying to understand: Does that mean that they're judging the confidence interval to be so wide that it's not meaningful? Clearly that would depend on what's being measured and the use case no?

#statistics #stats #rstats #stata #epi

New preprint: We present a new #dataset on the #German #language: DeReKoGram includes uni-, bi-, and trigram frequencies, lemma and POS information for a corpus of around 43 billion tokens.

We evaluate the distribution over the 16 datasets and present a (small) case study on #vocabulary growth.

At https://www.owid.de/plus/derekogram, we provide #Python, #Rstats and #Stata code that should help you getting started with the dataset.

Preprint available at https://doi.org/10.21203/rs.3.rs-3139640/v1.

#linguistics

Interested in #eusilc?
#SPSS, #Stata, and #Rstats routines to transform the raw data of EU-SILC into system files are now available for the most recent EU-SILC release 2023-release1: https://www.gesis.org/en/missy/materials/EU-SILC/setups

@officialmicrodata @gesis_gml @GESIS #eurostat

@Enema_Cowboy Yup! I teach a #Stata debugging workshop and part of it is about how to clearly explain the problem to someone else. Most times I think that resolves it! (For me, too)

Meetings with student collaborators where you pull up their code and check & debug it together take a lot of time, but they are uniquely satisfying. These are problems that can be solved within the span of a meeting! The student has wrestled with this and failed, and you are the fresh eyes they need! Something will be accomplished!

I'm an inveterate reader of advice columns, always have been, & I think this scratches the same itch: it is really fun to try to solve someone else's problem

#Stata

@b0rk Fun fact: Stata (extremely popular statistics software in the social sciences) defaults to 16- or 32-bit integers, spilling over to 32-bit _floats_ if your data gets out of the range supported by int32. This is so wild to me. It supports 64-bit ints and floats but explicitly makes them difficult to access.
#stata

#Stata package for sending emails from Stata with Powershell. Having used Stata for the craziest things, this is exactly down my lane https://docs.iza.org/dp16163.pdf

I have trained my whole life for this #parenting moment

#stata

You can really see the difference in aesthetics and #accessibility in a side-by-side of the same line chart in #Stata 17 (left) and 18 (right). Beyond cleaning up the grey background and dramatically improving the legend, the first 2 lines now have colors that I can easily tell apart despite my color impairment.

I appreciate all the analytic upgrades in #stata 18, but as someone with color vision impairment who could barely differentiate the first two default graph colors, I'm most excited about improvements to the color palette. #disability #data

#Stata 18 released!!! 🎉 👏 https://www.stata.com/new-in-stata/

Want to work with the #FReDA data, but need to learn how to use #Stata or just need a refresh? Sign-up for an introductory course hosted by GESIS - Leibniz Institute for the Social Sciences. The course will be taught by FReDA team members Irina Bauer & Annika Stein.

Sign-up here: https://training.gesis.org/?site=pDetails&child=full&pID=0x24F4A4F4987942088ED946FBAC6E803F&subID=0xF024CA6B216C46DB9C60748A0CED0B85

I've stumbled into reproducing a published analysis originally done in #Stata. As I'm diving (again) into the syntax, it strikes me just how error prone .do files are. Data wrangling seems to necessarily involve so much copy-paste and hand coding. There are options to commands that aren't in the standard documentation. You can abbreviate command names…up to a point.

I know #RStats syntax has its issues, and I'm trying not to mistake unfamiliarity for inherent problems. But whew this is a lot.

I have written a book draft of an introduction to #multilevel modeling, entitled #Multilevel Thinking: https://agrogan1.github.io/multilevel-thinking/. Comments, questions and corrections are appreciated, as are suggestions for a possible publisher.

While applicable to many different software programs, the book is currently centered around the use of #Stata, but I hope to extend it to use of #rstats (#lme4) and #julialang

I have built a tutorial on implementing multilevel models in #Stata, #rstats and #julialang: https://agrogan1.github.io/multilevel-multilingual/multilevel-multilingual.html .

@kaiarzheimer
those are books about econometric methods not about #stata, the mass would have been certainly the same with #rstats as support language, The book I used for inctroductory econometrics 30+ years ago was 600 pages with no mention of programming. if you cared about looking at least at the TOC, you could have seen there are quite a lot of things going on therein that completely justifies that number