Statistical Learning: 3.3 Multiple Linear Regression

Поділитися
Вставка
  • Опубліковано 29 січ 2025

КОМЕНТАРІ • 7

  • @PunmasterSTP
    @PunmasterSTP 23 дні тому

    Multiple linear regression? More like "Magnificent lectures to which you should listen". 👍

  • @zarinskie
    @zarinskie 5 місяців тому

    NOTES:
    In the equation, betas are unknown and X is observed to predict Y (known as predictors)
    Hyperplane is a plane that tries to minimize the squared distance between the points when you have multiple predictors
    p-values close to 1 are not significant
    completed 08.08.2024

  • @EBEALEX
    @EBEALEX Рік тому

    Thank you sir

  • @michaelcheung6290
    @michaelcheung6290 2 роки тому +1

    2:14 should it be not the closest point to the plane? As the closest point to the plane is the perpendicular projection onto the place but not f(x1, x2) in R^3

    • @sparse-manatee
      @sparse-manatee 2 роки тому

      The distance between a data point and its projection vertically to the plane along the vertical axis might be more easily interpreted.
      However, regression can also be solved using least-squares I think (which is in the sense of orthogonal projection you just mentioned).

    • @PunmasterSTP
      @PunmasterSTP 23 дні тому

      I think you're right; it should be the perpendicular projection, and I think the presenter probably just misspoke or didn't realize what he said wasn't accurate.

  • @OG-jz9fh
    @OG-jz9fh 10 місяців тому +2

    Hi, thanks for the info, but where are the codes? Without codes or the real examples of the statics in Python, it is just a dry class or session. Please, can we be like Karl Pearson? I am sorry to say it in a not very positive way. Thanks again.