Discover and read the best of Twitter Threads about #rstats

Most recents (24)

New variant tech briefing published today! assets.publishing.service.gov.uk/government/upl…

Today's R #tweetorial: The construction of this epicurve-in-an-epicurve, our first plot showing the increasing number of #DeltaVariant cases detected using sequencing AND genotyping. #Rstats #epitwitter
Genotyping with the reflex PCR involves looking for a combination of variant-specific mutations. Implementing genotyping surveillance has been a massive public health effort, and means a more complete, timely, yet alarming picture of Delta trends - but better we know!
To create the plot, we use ggplotGrob and annotation_custom, both ggplot2 functions. We combine an epicurve of daily Delta counts for the last 60 days with an inset of longer-term weekly counts. Here are the two ggplot objects seperately: [smallplot] (left) and [bigplot].
Read 8 tweets
1/4 In its first month, the free Epidemiologist R Handbook (epiRhandbook.com) was used 200k times by 32k people in 203 countries - thanks for sharing!

As a grassroots effort to improve access to R, we are now translating into 9 languages.

#rstats #epitwitter #appliedEpi Banner logo for the Epidemiologist R Handbook project. Two e
2/4 R software is a free, trusted tool for data analysis, but can be intimidating for new users.

The Epi R Handbook aims to enable use by public health practitioners on the ground, worldwide.

@paimadhu @Atul_Gawande @edyong209 @NickKristof @OWMorgan @EpiEllie @doctorsoumya
3/4 Written by and for applied epidemiologists, the Epi R Handbook is also available to download for offline use.

Developed with support of @tephinet (global network of field epidemiology training programs), #MSF, and the EPIET Alumni Network @EANBoard
Read 4 tweets
It's finally out, our (w/ @ricgallotti and @manlius84 ) work on the efficiency of parallel communication on complex networks w/ flows. nature.com/articles/s4200…
@ComuneLab @FBK_research Thread 1/
2/ This theoretical work started from a very real question on the structural and functional organisation of cities. epjdatascience.springeropen.com/articles/10.11…
When measuring the functional integration of cities through the flows of people within them, we stumbled upon a problem:
3/ If, on one hand, the communication efficiency was an established proxy for quantifying the integration of a complex system, on the other hand, there where different descriptors of efficiency for weighted networks!
Read 6 tweets
What do people in the EU feel most attached to - their region, their country, or Europe? Results from a survey of EU countries. #dataviz #RStats Image
A lot of people have wondered about the Basque Country and Catalonia in my map on attachment to regions. Do they not have strong regional identity? The answer is yes, but it is contested. There is large spread in the ratings. Image
Higher resolution version of the map of attachment to region, country and Europe (it is 5400*7200 in original): i.redd.it/0gy650wtd8371.…
Read 4 tweets
i'm diving into #rstats 💫tidymodels💫-the framework is incredibly powerful 💪, but there is a lot to learn! 🗻here are some resources i have found helpful (thank you developers and content creators!!)- please comment if you have others to add as well! 🙏
2019 A gentle introduction to tidy models by @theotheredgar rviews.rstudio.com/2019/06/19/a-g…
rstudio::conf(2020) Introduction to Machine Learning with the Tidyverse by @apreshill & @StatGarrett education.rstudio.com/blog/2020/02/c…
Read 8 tweets
New blog post up for #rstats ggplot custom 💫interactive💫 sunbursts!

🍰create some layers w/ geom_rect()
🔄 rotate w/ coord_polar()
🪄 sprinkle in some {ggiraph} for hover over interactivity 😎

demo data w/ #TidyTuesday salary survey

pipinghotdata.com/posts/2021-06-…
1⃣ interactive sunburst w/ variable length
2⃣ interactive sunburst w/ fixed length
Read 4 tweets
NEW: @Claire_Bushey & I analyzed allegations of police misconduct from three US cities (Chicago, NYC, Philly), and I have organized the #rstats DATA+CODE in a GitHub repo ⚡️⚡️⚡️

github.com/Financial-Time…

The @FT STORY:
ft.com/content/141182…

*THREAD* about the data
1/ To begin, here's an old-school chart from the Christopher Commission's independent investigation of the LAPD in 1991, after the police beating of Rodney King

The report found that the top 10% of officers w/misconduct complaints against them accounted for 28% of ALL complaints A chart from the Christopher Commission 1991 report showing
2/ ... over a 4yr pd. 30yrs later, this overall pattern—which the Commission called "remarkable and disturbing"—still holds true today in Chicago, NYC, & Philly

The 10% of officers named most often in civilian complaints against police drew ~ a third of all the complaints Bar charts showing the distribution of misconduct complaints
Read 28 tweets
Did you know you can build a complete 2D human neuroanatomical atlas from scratch entirely in #RStats, using @AllenInstitute data + some handy computational geometry, and plot structures using #ggplot? It's a fast-ish process too! Here my tutorial: github.com/langleylab/bra…
1/n
Among the things you can do: plotting structures of interest on the 3 axes, choosing different slices for each axis. Each structure is drawn independently and automatically. 2/n a plot of brain structures in the 3 anatomical axes, with ea
If you can match structures across datasets, you can plot other types of data such as RNA-seq gene expression from GTEx, BrainSpan or any other bulk transcriptomics dataset: 3/n a plot of GTEx normalized gene expression data visualized as
Read 4 tweets
Having trouble picking a color palette for your #Rstats visualization? Well here's a MEGA thread about all the ways you can choose a palette! 🧵[1/22]
We'll start off with the classics. First, you can use the palettes that are already a part of R's base code (keep in mind that these palettes were not necessarily designed to be colorblind-friendly)

[2/22]
The next classic set of palettes comes from {RColorBrewer}. You can use the command display.brewer.all(colorblindFriendly = TRUE) to see which palettes are colorblind-friendly.

(A guide to this package: datanovia.com/en/blog/the-a-…)

[3/22]
Read 22 tweets
Still responding to folks re: my transition to data science post! I'll get to everyone, promise!

Given the interest I thought people might want to know the (almost all free/low cost!) resources I used to train myself for a data science role

A (hopefully helpful) 🧵
R, Part I

My first real #rstats learning experience was using swirl. I loved that I could use it inside of R (rather than having to go back and forth between the resource and the RStudio console)

swirlstats.com/students.html
R, Part II

A cliche rec, but it's cliche for a reason. R for Data Science by @hadleywickham & @StatGarrett transitioned me from "kind of messing around" to "wow, I did that cool thing" in R. It's absolutely a steal that it's available for free

r4ds.had.co.nz
Read 14 tweets
1/5 Introducing The Epidemiologist R Handbook! Collaboratively written by and for applied epidemiologists. Check out epiRhandbook.com and share widely!
2/5 This free, open-access manual provides sample R code and tutorials addressing a wide variety of data management and visualization tasks with practical epidemiological examples.
3/5 The handbook is also available in an offline version for use in settings with low internet-connectivity, and is great for those looking to transition to R from SAS, Stata, SPSS or Excel.
Read 5 tweets
Atualização no Repositório Digital das Humanidades (PT-BR) para a @RedeSciELO

labhdufba.github.io/redhbr/scielo-…

Através de um conjunto de ferramentas é possível:

1) scielo_scraper: permite definir uma das 8 áreas de conhecimento e raspar os arquivos PDF ou apenas os arquivos XML;

1/4
2) scielo_ISSN: permite definir uma lista de revistas que serão raspadas através do ISSN. Assim como na ferramenta anterior, é possível definir o tipo de raspagem (PDF ou XML);

2/4
3) scielo_xml_to_csv: seleciona, organiza e salva as informações do dataset de arquivos XML das revistas previamente baixadas a partir das ferramentas anteriores. A ferramenta captura todos os metadados disponíveis dos artigos incluindo as referências bibliográficas

3/4
Read 5 tweets
SNEAK PREVIEW: Complete Path Enumeration (CPE) for CRM #dosefinding designs

Featuring a vignette for #Rstats package #precautionary update v0.2-3, currently under review by CRAN. 1/

precisionmethods.guru/DTAT/CPE/MCSE-…
This demonstrates the feasibility of a programme I’ve set forth, for the ‘exact’ simulation of ANY dose-escalation design.

Influenced by @MarkusTriska‘s exacting perspective on ’exactness’, however, I now use the more precise term ‘MCSE-free’. 2/
@MarkusTriska ‘MCSE-free’ in turn alludes to the title of recent work by Thomas M. Braun, from which I adapt my application.

(Braun’s mean-field approach truly can be called ‘simulation-free’, whereas I DO simulate the trials—just without Monte Carlo error.) 3/
Read 20 tweets
New preprint 📚🧠🧮🧒:
Oscillatory EEG Patterns of Arithmetic Problem Solving in Fourth Graders
osf.io/tmhce/
(1/9)
We already have a pretty good idea how arithmetic problem solving is reflected in oscillatory EEG activity in adults:
(2/9)
Simple problems are usually solved by fact retrieval. In that case, we observe theta ERS (power increase). Hard problems need to be calculated, which is reflected in alpha ERD (power decrease).
(3/9)
Read 9 tweets
Neste semestre, a @BeaMilz e o @jtrecenti ministraram a primeira turma do curso de Relatórios e Visualização de dados. Os trabalhos finais ficaram muito legais, e 3 foram premiados com uma bolsa de estudos em qualquer curso! Segue o 🧶:
#RStats #ggplot2 #TidyTuesday
Em ordem alfabética, o primeiro trabalho é da Ariane Hayana @arianehayana!

O trabalho foi um #TidyTuesday com dados sobre consumo de alimentos e emissões de CO2.

🖥️ Leia o trabalho em: curso-r.github.io/202103-visuali…

#rstats #ggplot2 #dataviz #DataScience
O trabalho da Brunna Escouto @brunnace foi um #TidyTuesday com dados relativos ao custo de obras de infraestrutura de trânsito, em especial metrôs, com informações de 54 países.

🖥️ Leia o trabalho em: curso-r.github.io/202103-visuali…

#rstats #ggplot2 #dataviz #DataScience #xaringan
Read 6 tweets
On Wednesday, I spoke to a group of academics and practitioners at @LboroSSEHS (via zoom) about some of the contemporary tools I'm using to make my evidence synthesis research easier & better.

If you're doing a systematic review or meta-analysis, here's a quick overview 👇
In terms of systematic review workflow, @Covidence is the best. Faster, easy to collaborate, online support, suitable for any research topic. Simplifies the process so much.

Cost is the only negative. Try your uni for a subscription!
Often with this type of research, you'll need to reverse engineer images of figures to extract the underlying data. WebPlotDigitizer (automeris.io/WebPlotDigitiz…) is a semi-automated tool that makes this process extremely easy.
Read 9 tweets
I recently implemented some pairs trading strategies for a paper, and decided to share an implementation of the Gatev, Goetzmann & Rouwenhorst (2006) strategy on a short article on RPubs.
rpubs.com/arubesam/Repli…
#rstats #RPubs #DataScience #finance #pairstrading #reproducibility
In the RPubs post above, I provide the #R code to backtest the strategy, as well as some results replicating Gatev, Goetzmann & Rouwenhorst (2006) and Do and Faff (2010), and extending the sample to the end of 2020. In this thread, I show some of these results.
Pairs trading is a type of systematic trading strategy based on finding pairs of stocks or assets that have historically "moved together", and betting that divergences will eventually get corrected. It is a simple form of statistical arbitrage.
Read 23 tweets
1/10 covidregionaldata 0.9.0 is now on CRAN! covidregionaldata is an #rstats package providing subnational & national #covid19 data in an open and robust framework. epiforecasts.io/covidregionald…
@seabbs @sbfnk @kathsherratt @RMartinNielsen @cmmid_lshtm @lshtm @RHULBioSci @BBSRC
2/10 The rewrite I have been leading over the last two months, as part of my PhD PIPS placement at @cmmid_lshtm, rewrites the package using an OOP R6 based approach and lays the groundwork for future extensions.
3/10 Released in April 2020, covidregionaldata was designed to process data needed for @sbfnk groups reproduction number estimates with the twin goals of making our tooling available for others and making our entire workflow #opensource and easy to review epiforecasts.io/covid/posts/gl…
Read 10 tweets
#Thread Comment des données en open data peuvent être le point de départ d’enquêtes journalistiques de terrain. Démonstration avec la base ARIA des accidents industriels gérée par le Bureau d’analyse des risques et pollutions industriels (Barpi) #DDJ #RStats 1/18 📊📈📉
La base ARIA est un inventaire des accidents industriels en France (pour la plupart) qui ont pu porter atteinte à la santé, la sécurité publique ou l’environnement. Elle n’est pas exhaustive. Mais elle peut être utilisée pour déceler des tendances et des similitudes 2/18
Cette base est en fait une vraie mine d’or. Pour chaque accident, elle donne la localisation, une description de l’évènement, et beaucoup de détails : l’environnement impacté, les matières impliquées, les causes de l’accident, le secteur d’activité, etc. 3/18
Read 18 tweets
ggiraph + shiny is fantastic!
~30 lines of code.
#dataviz #rstats
Part of today's interactive worksheet, see also here.
wilkelab.org/SDS375/schedul…
Read 4 tweets
`modelsummary` 0.7.0, the model and data summary package for #Rstats, is now available on CRAN. A thread on some of the new stuff. vincentarelbundock.github.io/modelsummary/
But first, here's a reminder that `modelsummary` supports 100+ statistical models; exports to Word, HTML, LaTeX, JPG, and PDF; and that its tables are *infinitely* customizable using packages like `kableExtra`, `gt` and `flextable`.
Plus, as a bonus, the package gives you a `modelplot` function for coefficient plots, and the crazy-flexible `datasummary` family of functions.
Read 11 tweets
1/10 Since October @ftargument has been running a crowd forecasting project for #COVID19. First in Germany and Poland and now in all ECDC nations + the United Kingdom.

You can contribute your forecasts here: cmmid-lshtm.shinyapps.io/crowd-forecast/ Image
2/10 We submit an ensemble of these forecasts to the ECDC forecast hub where they are combined with many other models from other forecasters and evaluated.

covid19forecasthub.eu/visualisation Image
3/10 This crowd forecasting method has been found to perform well compared to models submitted by multiple groups and to be one of the only methods to produce useful forecasts at longer time horizons.

medrxiv.org/content/10.110…
Read 12 tweets
Just gave an #rstats brain dump talk of pkgs/functions I've ♥️ over the last couple months. Aimed for the less obvious choices and called them "secret tRacks" after the cassette tape surprises (yes, I'm old) 🧵
`pak` rocks my world as package manger of choice over install.packages/devtools/remotes pak.r-lib.org
If you're not making animated svgs with `asciicast` you're missing out github.com/r-lib/asciicast
Read 15 tweets
(1/12) Announcing R CHARTS, the new #rstats and #rstatsES #dataviz site which includes, per language:

📈 Over 1100 #ggplot2 and base R charts
🧑‍🏫 Over 120 tutorials
📊 Over 50 different chart types
🎨 Several color & palette tools

r-charts.com

Made with #blogdown 🤩
(2/12) The site contains 8 different categories (the Spatial section is under construction) based on the Financial Times Visual Vocabulary:

raw.githubusercontent.com/ft-interactive… Image
(3/12) Once you pick a category you will see all the available #rstats charts for that category.

You can filter them by their chart type and base R or #ggplot2.

r-charts.com/correlation/
Read 12 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!