David Bernal

Our work with @DavidBernal224 and @gusgarciacruz on job spatial mismatch in Medellín has been published in the Journal of Transport Geography! Free access: authors.elsevier.com/c/1lUrz3RpEOeu…

We were thrilled to host the 2025 cohort of the Price Theory Summer Camp! Led by UChicago’s Kevin Murphy, the program brought together top PhD students from across the country.

Nuestro trabajo con @DavidBernal224 y @gusgarciacruz sobre desconexión entre trabajos y viviendas en Medellín está disponible como Documento de Investigación de @Banxico Documento: tinyurl.com/mr4r2wyp Resumen ejecutivo: tinyurl.com/5n9a94ff

Accepted :D

Is math discovered or invented? Most scientists say discovered. I used to agree. But I’ve come to think we invent systems to impose order, and math is our most powerful language for expressing that order.

Global reaction to Overleaf going down

Revise and resubmit :)

Paper here: jorgeperezperez.com/research/2020-…

@SolomonKurz @prisonrodeo @carlislerainey You can see the proof of the Gauss-Markov theorem in Fumio Hayashi's econometrics book. You do not need the assumption of normality to prove that OLS is BLUE. However, you need it to derive the distribution of b_hat- beta and use CLT.

@prisonrodeo @carlislerainey If I'm following this paper correctly, we don't have to assume the residuals are normal for OLS itself, but we do have to make that assumption (along with others) if we want OLS to contribute to the BLUE (best linear unbiased estimator).

When using OLS regression, from where do we get the assumption the residuals are normally distributed? You don't need that assumption to minimize the sum of the squared residuals, nor to use the CLT to compute the SE's for the beta coefficients. So what's its origins?