Why is this course important for academics? 

Academic researchers may find this course useful since methods of bayesian statistics are rarely taught as extensively as frequentist methods and this course may allow them to fill some of the gaps in their statistical training.

Why is it important for the labor market?

For applied researchers the bayesian approach offers a natural framework of the decision making process and familiarity with bayesian techniques often proves useful in data-driven decision making (e.g. launching a new feature in a tech company or implementing a policy in public sector). 

Description

In this class we are going to discuss the basics of bayesian statistics. While also rooted in the classical statistical decision theory, the bayesian approach rarely gets the same attention in a typical statistics / data analysis curriculum as the “frequentist approach.” The choice between frequentist and bayesian approaches often provokes heated debates. The origins and the current state of this divide is beyond the scope of this course. What’s often important in practice, is making good decisions using all of the information available and the bayesian approach is a natural formalism of this decision-making process.

>> Programe
 

Instructor

Nick Doudchenko
Ingeniero de Software
Google Resarch, Nueva York

Nick Doudchenko es ingeniero de software en Google Research en Nueva York. La mayor parte de su trabajo actual se centra en el diseño experimental y la estimación de los efectos causales en el contexto de los mercados en línea. Está particularmente interesado en combinar ideas estadísticas con técnicas de aprendizaje automático y optimización combinatoria. Antes de unirse a Google, trabajó como economista en Facebook, su primer trabajo después de completar un doctorado en Economía de Stanford GSB en 2018, donde fue estudiante de Guido Imbens y Lanier Benkard. Antes de eso, obtuvo una maestría en economía de la New Economic School y una licenciatura en matemáticas de la Universidad Estatal de Moscú.

Nick Doudchenko is a Software Engineer at Google Research in New York. Most of my current work is focused on experimental design and estimation of causal effects in the context of online marketplaces. I am particularly interested in combining statistical ideas with techniques from machine learning and combinatorial optimization. Before joining Google I worked as an Economist at Facebook which was my first job after completing a PhD in Economics from Stanford GSB in 2018 where I was advised by Guido Imbens and Lanier Benkard. Prior to that I received an MA in Economics from New Economic School and a BS in Mathematics from Moscow State University.

Formulario de postulación   Application form