GLM Part 5: Multivariate General Linear Models: Conditioning and Controlling

แชร์
ฝัง
  • เผยแพร่เมื่อ 6 ม.ค. 2025

ความคิดเห็น • 8

  • @mariavictoriabehler5754
    @mariavictoriabehler5754 ปีที่แล้ว +2

    OMG How much I like Statistics with this Videoooossss! Thank you... all the other people are so boring that I am not able to pay attention.

  • @sonja5058
    @sonja5058 3 ปีที่แล้ว +4

    Wow, I've taken 3 stats classes at uni and you're the first person that's made me really able to conceptualise residuals... thank you!

  • @brechjevanosch3863
    @brechjevanosch3863 ปีที่แล้ว

    Thanks for this! Can you tell me which is the video where you explain how to residualize?

  • @gpitwtei
    @gpitwtei 2 ปีที่แล้ว +5

    FYI you should differentiate between"multivariate" and "multiple" regression, which are very much not the same. In the title you have multivariate (=more than 1 DEPENDENT variable) but you talk about multiple regression (= more than 1 EXPLANATORY variable).

  • @CouttsDesign
    @CouttsDesign 3 ปีที่แล้ว +1

    Awesome vid! I'm wrapping my head around error correction. You mention that, "if you're in strict CDA territory, you can compute p-values for those values that you're interested in." I'm using logistic regression with SPSS. Does that mean I can just ignore the p-values for variables that I'm not interested in and just report the p-value for the one I am interested in? Or is there another computation you're referring to?

    • @QuantPsych
      @QuantPsych  3 ปีที่แล้ว +1

      Yep, that's what I mean.

  • @pairotechakranon1568
    @pairotechakranon1568 3 ปีที่แล้ว

    I love all your materials..awesome. Do you have any online courses or available online materials in which I could attend your classes? It's really appreciated and I am willing to pay for your tuition fees, such as Udemy etc. you are the only teacher who I admire and inspire me to learn all about statistics :)

    • @QuantPsych
      @QuantPsych  3 ปีที่แล้ว

      Thanks! The trouble is finding the time :) Maybe someday.