When I say that Claude is very good at coding in R, there are caveats.

As G10 countries succumb to simplistic populism, reckless fiscal policy and fiscal dominance of their central banks, the line between the G10 and EM is getting blurred. Real long-term interest rates in the G10 have begun to rise towards EM levels... robinjbrooks.substack.com/p/the-global...

Closeread Prize winners are here! 🏆 Want to create engaging, scroll-based stories with your data? See stunning scrollytelling examples made with Quarto: posit.co/blog/closere... Learn more about Closeread and its power to create dynamic narratives. Congrats winners! #Quarto #RStats #Python

Claude can be savage at times

What happens when you get lazy and don't follow your own advice...

Happy to know I'm not the only one who creates a ggplot2 theme package for every company I work with #positconf. Not only does it standardize communication, but it also makes a great impression. theme_quintoandar() took about a day to make and led to me reporting directly to the C-level.

Been working with ClaudeCode for the past ~2 months (mainly R and Quarto). My main tips: 1. Plan a lot. 2. Manually approve (auto-accept only for very simple tasks). 3. Create knowledge that can be referenced. 4. Avoid /compact and auto-compacting.

I've been using Air since switching to Positron and have been loving it. My only criticism is the styling of tibbles (it just breaks them). Apparently that has been solved # fmt: tabular coming soon #PositConf

The fastest way to start using & exploring the new 2024 1-year ACS data? R with tidycensus + mapgl. With just a few lines of code, you can pull down any of the thousands of variables available in the new data and explore on an interactive MapLibre map.

Jair Bolsonaro was sentenced to more than 27 years in a Brazil prison for attempting a coup after losing the 2022 election.

I was loving Claude Code... until I tried it with #rstats. Constant errors, wouldn't use the tidyverse even when asked, "optimized" functions were slower. Frustrated, I started a session just to teach R to Claude and summarize what it learned into a CLAUDE.md file gist.github.com/sj-io/3828d6...

Saw a post from an influencer telling followers to "stop using R for anything -- use Python like a normal person" As a 10+ year R & Python user, the irony is that 2025 is the best time EVER to be using R. Here's why:

What a great piece on online advertising. Summarizes beautifully my frustrations when working as a data scientist (with econ background) in a big marketing department. Leaders justified decisions with bad metrics, fueled by biased samples. thecorrespondent.com/100/the-new-...

Forgot some Positron shoutouts, @andrew.heiss.phd has a nice post talking about Positron (www.andrewheiss.com/blog/2024/07...) and Albert Rapp also has one (3mw.albert-rapp.de/p/using-the-...)

Almost reaching 100 posts in my blog restateinsight.com/blog. Feel like it's time to reorganize things. I created this blog back in 2019 to post R tutorials on econometrics and data visualization. I posted everything in Portuguese since I felt there was lack of R-related materials in Portuguese.

Overall, Positron has some great benefits over RStudio. It's faster, lighter, and after some customization looks a lot nicer. There are still issues. The main one that comes to mind is the overall inferior autocomplete. Also, autocomplete for Quarto docs seems to breakdown at some point.

Last month I tried to pick up Positron as my main IDE. First two weeks were rough and I found myself going back to RStudio some times for both Quarto and R. After much stress I am now using Positron daily. The process did make me realize how much of a petty person I am when it comes to IDEs.

City of London: - 57% more cycling since 2022 - 1.8 times more bikes than cars - People walking, wheeling and cycling now make up 75% of observed travel activity - Motor vehicle volumes down 70% since 1999 www.cityoflondon.gov.uk/assets/Servi...

Economic contraction in the US is attributable entirely to tariffs

Still missing some plots, but wanted to share this one. Day 25 #30DayChartChallenge | Uncertainties: Risk The plot shows the "Brazil Risk" measured by the EMBI+ and highlights key economic/political events. This index broadly measures the risk that foreign investors face when investing in Brazil.

Day 30 #30DayChartChallenge | National Geographic This chart shows the deforestation of the Amazon rainforest over the years, based on satellite imagery. This year's COP30 will take place in Belém, Pará—a state that has consistently ranked among the top 2 in deforestation, since 1988.

...and he's doing this with his party in control of both houses of congress www.ft.com/content/9d3e...

Day 29 #30DayChartChallenge | Uncertainties: extraterrestrial For day 29, I aggregated U.S. box office numbers for movies featuring aliens—any kind of extraterrestrial life counts. Alien sci-fi really dipped between 1986 and 1995 (Predator 2, Alien 3, Coneheads...).

Day 4 #30DayChartChallenge | Comparisons: Big or Small Choropleth maps of Brazil can be very deceiving. The Midwest and North regions cover about 65% of the country’s area but house only 16% of its population. Maps show Aging Index: ratio of the pop. aged 65 and over to those aged 14 and under.

Day 27 #30DayChartChallenge | Uncertainty: Noise For day 27 I present several plots describing statistical properties of white noise.

Day 15 #30DayChartChallenge | Relationships: Complicated For day 15 I show the growing gap between rental prices and sales prices in Brazil. As of February 2025, house prices, on average, have increased 58% more than rental prices. This phenomenon is seen across several major Brazilian cities.

Day 13 #30DayChartChallenge | Relationships: Clusters The map shows the number of job opportunities accessible within a 30-minute drive during peak hours. The data comes from the AOP project (github.com/ipeaGIT/aopd...)

Day 11 #30DayChartChallenge | Distributions: Stripes An updated version of the classic Ed Hawkins global warming stripes.

Day 10 #30DayChartChallenge | Distributions: Multi-modal Cars accidents are usually strongly correlated with rush hours. Even though, there's the usual early morning and late afternoon peaks, it feels like it's always rush hour if there's daylight in São Paulo.

Day 7 #30DayChartChallenge | Distributions: Outliers Wealth inequality is very high among Brazilian cities. Maricá, an oil royalty recipient, is 74 times wealthier than the poorest city on the list.

Day 26 #30DayChartChallenge | Uncertainties: Monochrome Economic recessions have decoupled electric energy consumption and real GDP in Brazil. A big gap appeared after the 2014-16 economic crisis.

Day 9 #30DayChartChallenge | Distributions: Diverging The charts show the geographical disparities in illiteracy rates across Brazil. Overall, cities in the Northeast show significantly higher rates than those in other regions.

Day 3 #30DayChartChallenge | Comparisons: Circular Do people live near downtown? I estimate the number of people living within a 6 km radius of their city's downtown. The street colors represent nearby population density: darker shades indicate lower values, lighter shades represent higher.

Day 19 #30DayChartChallenge | Time-series: Smooth This plot shows the daily closing price of the S&P500 Index since October 2024 with several different smoothing and trend estimation techniques. Shoutout to base R (lots of convenient smoothers in stats) and TTR.

Day 20 #30DayChartChallenge | Time-series: Urbanization The rise of solo-living is a global trend and Brazil is following suit. Single family households have been increasing in all states. Currently, 18 million live by themselves in Brazil.

Day 21 #30DayChartChallenge | Time-series: Fossils For day 21, I chose to make a plot showing the continued rise of global fossil fuel consumption.

Day 22 #30DayChartChallenge | Time-series: Stars The plot shows sunspot cycles and the contraction cycles of the US economy in the past 150 years. The plot was made with ggplot2 and R.

Day 23 #30DayChartChallenge | Time-series: Log-scale Economists really love their logs and this love is justified. Despite the nasty log(x + 1), applying a logarithmic transformation helps modeling in too many ways to list. For most macroeconomic time series, logs are the way to go.