Bob

Antes de começar pelos avançados, deixo como sugestão de nivelamento um clássico. Courant e Robbins de 1941 é um elegante convite à matemática. Começando por teoria dos números, passando por algebra, geometria, topologia até o cálculo. É uma referência eterna.

By now, I have published a fair number of papers, and one more acceptance would have close to zero marginal impact on anything that matters professionally. But getting my survey on “Deep Learning for Solving Models” accepted into the Journal of Economic Literature made me genuinely happy, for reasons that have nothing to do with my CV. I had the misfortune of studying my undergraduate degree in economics at a quite awful institution. Two professors, David Taguas and Alfredo Arahuetes, were outstanding, and I owe them a great deal. The rest were well below any reasonable professional level, and some violated the basic standards of ethical conduct. They had no business teaching economics at any level, let alone at a university that charged tuition and claimed to prepare students for professional life. I had to work out most of my education on my own. The surveys published in the Journal of Economic Literature were how I did it. I spent hours in the library’s reading room going through one survey after another on topics I had never been properly taught. Some helped more than others, but collectively they gave me a solid enough foundation that, when I arrived at Minnesota for my PhD, I discovered, to my considerable surprise, that I was ahead of nearly all the other first-year students, including some who held master’s degrees, despite the fact that I had finished my undergraduate degree just six weeks before. I owe the Journal of Economic Literature a debt I will never be able to repay. Publishing a survey there is the closest I can come to trying. So, the thought that some student somewhere, working on her own in a library or on a laptop, might find my survey useful gives me tremendous satisfaction. But there is a broader point worth making. Even in the world of AI, the profession has an important mission in making educational material widely available. Textbooks, surveys, teaching slides, these are public goods in the economist’s sense: high social value, insufficient private incentive to produce. This is also why I post all my slides and teaching material online: sas.upenn.edu/~jesusfv/deepl… We do not reward these activities nearly enough, and their supply is well below what any reasonable social planner would choose. I do not have a good proposal for changing this, and I would welcome suggestions. What I do find heartbreaking is that many of the great economists of the past couple of generations never wrote textbooks on their areas of expertise. I do not mean this as criticism. All of them maximize, and perhaps they all suffer from the same bias I suffer from: the belief that one can always do it next year. But I often think about the hours of pure intellectual pleasure I would have had reading “Time Series Econometrics: An Advanced Textbook” by Chris Sims or “Methods in Structural Estimation” by Pat Bajari. Those books do not exist. They should.

@danveloper We gonna have a Mac Studio (M5 ultra 512GB) for $10k more powerful than a $100k GPU server by like a month from now.

We’re gonna see someone running a 1T model at 100tok/s on a $2500 laptop by like a month from now. Long Apple, this is the best AI inference platform.

@femisapien_z Otimo para perder dinheiro sem esforco.

@krunkosaurus you should be able to run the qwen 3.5 122b in ollama with this hardware.

Apple M5 Max Macbook Pro with 128GB ram arrived. First unoptimized chat with Qwen3.5 is 108 TPS! Local AI is here.

An Homage to Christopher Sims: We built our cathedrals on the holy trinity of Consistency, Efficiency, and Unbiasedness. We swore that the data, and only the data, would speak. We hunted the Null Hypothesis with the ferocity of inquisitors, armed with nothing but our p-values and a refusal to admit we had any expectations at all. But then came the Macro-Reality. We found ourselves in a desert of "Incredible Restrictions." To make our models work, we were torturing the data into structural boxes that didn't fit, just to keep our Frequentist "objectivity" intact. Our Vector Autoregressions were beautiful, but they were dying of thirst: too many parameters, too little time, and a covariance matrix that looked like a crime scene. Then came Sims. He didn't just walk into the room; he redefined the architecture. He looked at our "unbiased" estimates, spinning wildly into the infinite, and showed us the Minnesota Prior. To a die-hard Frequentist, it felt like a betrayal. He was telling us to "shrink" our estimates toward a random walk. He was telling us to admit, mathematically, that we already had a hunch about where the economy was going. It felt like flirting with the dark side: replacing the "purity" of the Likelihood with the "convenience" of the Posterior. But here is the tribute: Only a giant like Sims could make the Dark Side look like the Enlightenment. He didn't use Priors because he was lazy; he used them because he was honest. He realized that an "unbiased" forecast that is 1,000% wrong is useless, while a "biased" Bayesian forecast that actually hits the target is statesmanship. He turned the "Prior" from a confession of ignorance into a tool of surgical precision. So, we tip our hats to the man who forced us to admit that even in the strictest Frequentist lab, the ghost of the Prior is always in the machine. He didn't break the rules; he just showed us that the rules were keeping us from the truth. In the end, Sims taught us that the Prior justifies the means.

Vocês que sabem de tudo, qual é o melhor modelo de IA atualmente?

@agpatriota @MatInvest1 Exato e nem não se adaptar com as novas ferramentas vai ficar para atras, mais rapido que imaginam.

Se todos têm acesso a ferramentas que aumentam a eficiência, a régua muda. Alguém vai usar o tempo para fazer mais e se tornar super produtivo em relação a quem não usa. Quem não adota o mesmo comportamento fica pra trás. Em algum momento, o sistema encontra um novo equilíbrio e a maioria passa a trabalhar mais.

Já repararam que, quanto mais a tecnologia se desenvolve, mais horas por semana as pessoas precisam trabalhar para sobreviver? Não era para ser o contrário? Onde está o golpe?

@ArthurMotaSP @claudiolucinda uso para tudo, até mesmo como chat. O grosso é codigo, mas muita apresentação e documentos. O melhor professor é o proprio CLI. O que estou gostando no statistica code é já tem todos plugins e dados nativos para economia e financas e o CLI sabem bem o que com tudo.

@claudiolucinda @rationalexpec como estao usando? dicas de lugar para aprender mais? tenho feito coisas mto basicas e que ja fazia no gemini

@claudiolucinda @ArthurMotaSP usar o chatgpt inves do CLI é como usar excel inves de python. CLI a sua imaginação e criatitivade é o limite.

@miniapeur claude code with academic skills.

I have a draft of a paper, and I am thinking of using a chatbot to review it (grammar, general flow/structure, mistakes, and suggestions). Does anyone have experience with this, and which chatbot is best for this?

@claudiolucinda statistica code (que usa gpt-5.4-xhigh) mas é voltado para empresas.

@DAcemogluMIT I will ask Claude to summarize

Dear followers, I’m happy to share this new academic paper on how even capable AI can lead to deterioration of collective knowledge in society

Acompanhe a primeira live do BDM, nesta terça-feira, às 9h! O que COPOM vai fazer? Com Sérgio Goldenstein, ao vivo! youtube.com/live/KCkcXQjus…

@goescarlos "spanning" até mercados completos seria uma otima noticia. Mas está longe de ser o caso num pais amante de loteria.

@rationalexpec Arrow Debreu securities chegando no Brasil, bob

Otimo. O Brasileiro vai ser mais um forma de perder dinheiro no final de semana.

@nimblesprocket @leticiaarsenio não tem estratégia que de 2% ao dia com consistencia, mesmo com 90% acerto. nao tem hedge fund nem prop trading que dá isso. Se voce fosse trader que opção (como quer tirar onda) saberia disso.

@leticiaarsenio @rationalexpec É uma estratégia de opções. Fala aí pra mim agora sem pesquisar pra que servem as 5 gregas em trading de opções e o que é gamma scalping

O filantropo descobriu a mina de ouro e tá dando de graça só por uns likes. (2% ao dia voce engole o Elon Musk em poucos anos comecando com $1000).

@leticiaarsenio qualquer promessa de retorno acima do CDI (ou treasury no caso) sem risco é golpe. Se for acima tem risco e quando mais acima maior o risco. Não existe magica nem almoço gratis.

@rationalexpec 2% ao dia é mais mentiroso do que o CDI do Banco Master

@miniapeur good, but I prefer uv

Are you guys using Poetry in Python? If so, why?