I am teaching an introduction to mathematical statistics and yesterday I introduced Bayesian inference. In one of the examples I elicited a prior distribution to the probability of Neymar be announced as a member of the Brazilian squad for the 2026 FIFA World cup.

Even for a Bayesian in an ML/AI new data science world, I find quite pleasure to prove in classroom that an MLE, under smooth conditions, is consistent and asymptotically Gaussian.

I still prefer studying a subjective, specially new subjects, using printed version books. Unfortunatelly it is quite expensive buying them from Brazil. And the most cutting edge is a subject the harder it is to find a book on that subjective in a library here. This is science inequality.

I remember last year results, some news were making fun of the android performance (the fasted one was 2h 40'). I was impressed since my aim in a half-marathon is 2hs and some androids did finish a half-marathon. Then just one year later, an android beat the human record (56') by 6 minutes! Wow!

The research Joint Bayesian Nowcasting of Severe Acute Respiratory Illness and COVID‐19 Positives in Brazil is published in Statistics in Medicine. doi.org/10.1002/sim....

My research, "Mortality trends for diabetes mellitus, hypertension and cardiovascular disease among people living with and without HIV in Brazil during the COVID‐19 pandemic, 2020–2022", has been published! Read it here: doi.org/10.1111/hiv.... #publishedarticle @wileyinresearch

Very intense day: - Paper accepted in the morning (and one rejected) - Student defending her PhD (she did a wonderful job) - Final report of a project submitted - Scientific initiation grant (PIBIC) submitted - Paper awarded in a event in Brasilia

Two postdoc positions for a number of exciting epidemic modelling projects opening in my Unit at @pasteur.fr in beautiful Paris. Deadline for applications: 13th April. More details at the following link: research.pasteur.fr/job/two-post...

It is unsafe for other Scientific meetings too. My favorite meeting is the @isba-bayesian.bsky.social, this year will be in Japan (unfortunately I won't go for personal reasons) and the next one in 28 will be in the US which I won't go either because I am not keen to pass through the visa process.

Socó-dorminhoco (Nycticorax nycticorax)

Our paper on the Infodengue-Mosqlimate challenge to predict the next dengue season in Brazil is out in @pnas.org www.pnas.org/doi/10.1073/...

I thought that during the course of these 15+ years after the PhD I would be less annoyed by a manuscript rejection...

The orange man always surprises me. I do expect the worst from him, but somehow he exceeds himself. I have no words for this 💩

End of holidays and after the kids sleep usually while cooking or doing the dishes I watch animes (Japanese cartoons). An old habit that I have kept doing since a teenager. Recently I have been following these: - Frieren - Sentenced to be a hero - Camping solo for two - Touring after the apocalypse

If you put Greenland over Brazil. It is a massive piece of land, but not as massive as one would believe by looking at a Mercator projection.

A very short guide to Bayesian reasoning www.bbc.com/videos/c78ee...

Lula should decline the invitation to join Trump's board of peace. Brazil always wanted to be a permanent member of the UN's Security Council. But this Trump's private version is surreal as most of the things the orange man does and says...

"If Brazilians submitted a shoe to the Oscar, everyone would vote for it, says the Sirât's director" Really? How many awards for best international feature film has Brazil won? Well, just one in six nominations... I hope we get a second one.

This is indeed the best meme song of 2025. I love it m.youtube.com/shorts/fa3U-...

Here in Brazil all graduate programs are evaluated by a government agency called CAPES, and the programs get grades up to 7 every 4 years. For reference, PhD programs usually start with 4 points. Programs with a 7 are usually internationally recognized by peers.

2026 resolutions: - starting running again (I had a knee injury just before the pandemic, then I stopped doing sports running and taekwondo) - camping / hiking (alone and/or with family). There are some great places near Rio, like the Parnaso - work less (I was this close to a burnout this year)

I have recently realised that I am not an intellectual person, some may call me a scientist which I am not sure if it is right. I am Brazilian civil servant (public health researcher) and a part time lecturer devoted to (Bayesian) stats mostly applied to infectious disease epidemiology.

No Rio de Janeiro temos um mês inteiro. 😅

I need this sort of holidays. Translating: - Stop thinking about work. - Sorry.

I was going to reject reviewing a manuscript. I entered the journal website and I thought I was accepting the cookies of the journal website but I was in fact accepting the revision 🤡

SARI (Severe acute respiratory illness) cases with influenza detection in Brazil. We had a clear outbreak associated with H1N1 and some states had a second outbreak associated with H3N2. This second "wave" different from the first one is completely out of season.

Yesterday, I gave a talk on the E-vigilancia ("e-surveilance") conference at the FGV (Getúlio Vargas Foundation). My talk was about how Google trends data could improve nowcasts and short term forecasts of dengue cases in Brazil. Marcelo (Fiocruz) and Flavio (FGV) were in the same session.

This is somehow scary. Everyone is aware that AI has been used during the writing of a manuscript, but as expected it will eventually be part of the revision process. So scientific publication is going through a big change, and I guess the impact of research will no longer be measured on papers.

The linkedin games for some reason stopped working on my mobile so I lost my sequences. Although it is only a couple of minutes per day,I haven't realised I was consistently wasting plenty of time with them. I'm glad I am free of them...

Since I've been working on statistical modelling of infectious diseases from the pandemic onwards I got involved in too many projects leading to several publications which seems nice but it is not! If I could go back in time I would definitely work less, half of it or less. I am exhausted now!

In this paper, we describe the southernmost dengue outbreak ever reported. The outbreak occurred on the city of Bahia Blanca, Argentina (~600km south of Buenos Aires). www.sciencedirect.com/science/arti...

NEW PAPER We developed a statistical model to forecast dengue cases for the next season using the history of cases only. We suggest using the posterior predictive to define epidemics bands. For instance, we could see how huge was 2024 dengue season in Brazil. www.sciencedirect.com/science/arti...

📢 NEW PAPER! In response to 2024 dengue epidemic we built Dengue-tracker 🦟📈 to improve nowcasts in Brazil by combining surveillance data + Google Trends 🔗 diseasesurveillance.github.io/dengue-tracker 📄 doi.org/10.1371/jour... 🙌 With Guilherme Soares,Rafael Izbicki,Yang Xiao&@leosbastos.bsky.social

I usually don't like pictures of myself, but I like this one.

When Thomas Bayes meets John Snow. Presented in Vitória, capital of Espírito Santo state, at the 69th RBRAS meeting and 21st SEAGRO

The #JSM2025 is about to start, the JSM is by far the biggest stats conference I've attended. I went to Miami Beach and Boston, it was a great experience. But nowadays I prefer relatively smaller conferences and outside the US (my visa expired and I am not keen to pass through the application again)

In two weeks, I will be a keynote speaker in a national stats conference. I am quite happy about that but also very nervous. I will talk about Bayesian learning on infectious disease surveillance, so it is time to start preparing my talk (and finish it 10 minutes before the actual conference... 😅)

Can you imagine if last year the Brazilian visa were revoked from the NY judge and the jury members (and their families) after Trump being convicted of crimes? Super weird, isn't it? Well, it seems the opposite is OK for some... www.theguardian.com/us-news/2025...

Trump will soon add new tariffs for all countries to guarantee global security like Italian mafia or Brazilian militia does on their territories (you can name your "favourite" group/organisation/gang/whatever that charges others to guarantee their security)

I am quite surprised and impressed by Chelsea's result at the final against PSG. But the match I am interested in today is from another blue jersey team at the Brasileirão 🦊😅😬 Cruzeiro X Grêmio

Many thanks @leosbastos.bsky.social and everyone at Fiocruz for the warm welcome! I really enjoyed hearing about all the exciting work happening here and visiting the beautiful Rio de Janeiro! 💚💛💙 #rstats #GISchat #diseasesurveillance

It is a bit frustrating knowing that the 2028 ISBA World meeting will be in the US (Milwaukee). Nothing against the venue itself, but to have to pass through the annoying Visa process a Brazilian has to do. It may be convenient for most Bayesians, but it is not for me.

We had a very special visitor today, @paulamoraga.bsky.social gave us a great lecturer on Spatial data science in R. Thanks Paula!

The manuscript "Duration of COVID-19 symptoms in children: a longitudinal study in a Rio de Janeiro favela, Brazil" was recently accepted in the BMJ Open. I am one of the authors and, once published, it will be my 100th peer reviewed paper. A meaningless academic milestone, a cool number though.

Europeans: we don't really care about the FIFA world club... Dude, my team is not even there and I am watching the games even supporting some clubs (the weaker ones which maybe are no longer the Brazilian ones...)

As well as interesting data, nice piece of data viz to have outlier line stretch beyond y axis…

The Fiocruz main campus is at Av. Brasil. I feel sad to see it called "avenue of Death" but it is true... 😥 www.theguardian.com/world/2025/m...

Save the dates and come to Rio to talk about Data Science in Public Heath, epidemiological surveillance and the impacts of the climate change. fiocruz.tghn.org/news/e-surve...

Today (25/5) is Fiocruz 125th year anniversary, instead of the iconic castle picture (which turns out I have no one in my phone) I post a picture of the cell from the Life's museum which is part of the Fiocruz campus in Manguinhos and I place I often take my kids to play around there.

Our (@leosbastos.bsky.social‬ and I) comment about dengue risk is highlighted in the cover and the cover itself!!! Link to our comment: www.thelancet.com/journals/lan...