The hottest Quantitative Methods Substack posts right now

And their main takeaways
Category
Top Science Topics

Quantifying Curriculum Degradation: Evidence from the University of Chicago

Heterodox STEM • 192 implied HN points • 15 Mar 26
🚌 Education Quantitative Methods
  1. A keyword-based method can flag courses as engaging with progressive ideas or the Western canon, and while this approach is blunt and prone to errors or manipulation, it is useful for tracking changes over time and comparing institutions.
  2. At the University of Chicago (2012–2025) the share of courses matching progressive keywords rose from about 12.7% to 28.3% while canon-related courses stayed near 12%, so progressive signals now outpace canon signals especially in humanities and social sciences and even show up in STEM.
  3. A public Curriculum Content Index built from catalogs, syllabi, and enrollment could give families, donors, and policymakers transparent comparisons across universities, but such an index should be treated as a noisy first pass and not as a basis for micromanaging curricula or replacing careful evaluation.

1. Three Rules for AI in Finance

Arpitrage • 2299 implied HN points • 02 Feb 26
đź’° Finance Quantitative Methods
  1. AI creates simpler, lower-dimensional maps of a complicated world so people can act on it; judge models by whether they improve real decisions and the cost–quality tradeoffs, not just narrow benchmarks.
  2. AI gains are capped by the slowest bottleneck in a process (Amdahl’s Law), so focus on speeding up the truly constraining steps — often regulatory, organizational, or incentive-related rather than purely technical.
  3. Automation drives prices down for commodified tasks and raises the value of scarce complements like private information, relationships, and judgment, so follow price signals and elasticities to see what gets automated and what stays valuable.

Quantitative Long-Run Global Economic History: Econ 196: Seminar: Special Topics in Economics (Spring 2026)

Brad DeLong's Grasping Reality • 123 implied HN points • 21 Jan 26
🚌 Education Quantitative Methods
  1. The course is a quantitative, long-run tour of global economic history covering everything from early humans and the rise of agriculture to industrialization, globalization, and modern attention/info/biotech economies, with a focus on causes of growth, inequality, and institutions.
  2. The pedagogy stresses hands-on data-science methods—sampling, estimation, forecasting, simulation, and counterfactual modeling—designed to let both humanists and quants learn to model parts of the world economy without prior coding experience.
  3. There are firm expectations: mandatory pre-class readings and a short assignment answering five questions (including on using AI/LLMs), and prompt submission is required to shape the next class session.

Data in, garbage out

Hypertext • 0 implied HN points • 27 Mar 24
🔬 Science Quantitative Methods
  1. Statistics can only tell us so much, so we should approach data with humility about both the power of social programs and hard data to test them.
  2. Rigorous measurement often doesn't definitively show whether interventions work, leading to ongoing debates and conflicting results in various fields.
  3. While randomized controlled trials have their value in measuring specified outcomes, they can miss unexpected effects and subtle interactions, highlighting the importance of qualitative methods and personal observations.

Econ 196: WEEK 1: Introduction, & the Very Longest Run Shape of Human Population History

Brad DeLong's Grasping Reality • 0 implied HN points • 02 Jan 26
🚌 Education Quantitative Methods
  1. The course is a quantitative, long-run global economic history class that teaches data-science literacy (including Python) to analyze population and income trends.
  2. Grades are intentionally generous but contingent on showing up, doing pre-class work, and participating—skip or zone out and you lose that privilege.
  3. Expect weekly short writing assignments, background readings, small data exercises, and optional Thursday Zoom sessions, with all logistics and materials posted on the course site.
Get a weekly roundup of the best Substack posts, by hacker news affinity: