|
This course introduces students to analyzing and
employing machine learning algorithms to evaluate
public policies. To that end, the student first
becomes conversant with the core issues of causal
statistics, such as the potential outcomes
framework, drawing causal diagrams, and
recognizing sufficient conditions for statistical
identification. Simultaneously, the class touches on
the building blocks of R, including data wrangling
and functional programming. After acquiring basic
knowledge of coding and causal statistics, the
material gravitates around the building blocks of
machine learning (ML) and their implementation in
R. Subsequently, the student learns about the
meaningful overlaps between causal statistics and
ML by reviewing the notions of Causal Trees and
Causal Forests. Finally, a significant portion of the
course addresses a series of applications
concerning evaluations of public initiatives, such as
police reforms, environmental preservation, and
educational programs.
|
