Skip to content

Latest commit

 

History

History
8 lines (5 loc) · 539 Bytes

unit_04_part_2.md

File metadata and controls

8 lines (5 loc) · 539 Bytes

Unit 4: Part 2

Best Predictors (and their Linear Approximations)

The second part of this week's work introduces two of the core ideas in the course:

  1. With data, we can produce conditional statements that are very good summaries of an underlying joint distribution function.
  2. We can require that our conditional statements have a relatively simple form—linearity in the inputs—and show that in some cases these very simple functions can still serve as a very good summary of the underlying joint distribution function