GLM: Iteratively Re-weighted Least Squares for a General Link Function

แชร์
ฝัง
  • เผยแพร่เมื่อ 16 ม.ค. 2025

ความคิดเห็น • 28

  • @grantlawley338
    @grantlawley338 2 ปีที่แล้ว +1

    This is a much better explanation than any article or paper I have read on the subject. Thanks for making the video!

    • @statisticsmatt
      @statisticsmatt  2 ปีที่แล้ว

      I absolutely love hearing that. Makes my day. Many thanks for watching. Don't forget to subscribe and let others know about this channel.

  • @nhlanhlamsongelwa4364
    @nhlanhlamsongelwa4364 4 ปีที่แล้ว +1

    just a few minutes in and you've already cleared some things I wasn't completely getting. thank you for putting this series together.

    • @statisticsmatt
      @statisticsmatt  4 ปีที่แล้ว

      Glad to hear it, this makes my day! Many thanks for watching!

    • @nhlanhlamsongelwa4364
      @nhlanhlamsongelwa4364 4 ปีที่แล้ว

      @@statisticsmatt do you have any videos that cover log likelihood ratio statistics by the way?

    • @statisticsmatt
      @statisticsmatt  4 ปีที่แล้ว

      @@nhlanhlamsongelwa4364 Not yet. But this is definitely on my list to do. Unfortunately, it may take a few months for me to complete. Many thanks for watching!

  • @warrenpettine7574
    @warrenpettine7574 4 ปีที่แล้ว +1

    Thanks for putting this together! Really helped.

    • @statisticsmatt
      @statisticsmatt  4 ปีที่แล้ว

      You're welcome. Many thanks for watching!

  • @chenxiliu7376
    @chenxiliu7376 4 ปีที่แล้ว +1

    Fantastic! It helps a lot!

    • @statisticsmatt
      @statisticsmatt  4 ปีที่แล้ว

      Many thanks! And thanks for watching!

  • @igoramelichev6664
    @igoramelichev6664 ปีที่แล้ว

    Thank you!

    • @statisticsmatt
      @statisticsmatt  ปีที่แล้ว

      You're welcome. Many thanks for watching. Don't forget to subscribe and let others know about this channel.

  • @mateoordonez9449
    @mateoordonez9449 ปีที่แล้ว

    This might have save my up-coming exam on Generalized Regression. Thanks!

    • @statisticsmatt
      @statisticsmatt  ปีที่แล้ว

      I love hearing that the videos are helpful. Many thanks for watching! Don't forget to subscribe and let others know about this channel.

  • @pan19682
    @pan19682 5 หลายเดือนก่อน

    😅 what an amazing awesome presentation thanks a lot professor 🙂

    • @statisticsmatt
      @statisticsmatt  5 หลายเดือนก่อน

      I love hearing that the videos are helpful! Many thanks for watching. Don't forget to subscribe and let others know about this channel.

  • @joshdavis5224
    @joshdavis5224 ปีที่แล้ว

    When you refer to BV4 at 25:50 what video is that?

    • @statisticsmatt
      @statisticsmatt  ปีที่แล้ว

      First, I'm loving the questions! I'm on vacation for a couple weeks and upon return, I'll try to answer these questions. Many thanks for watching! Don't forget to subscribe and let others know about this channel.

    • @statisticsmatt
      @statisticsmatt  ปีที่แล้ว +2

      @joshdavis5224, at some point I re-watched the playlist and didn't like BV4, so I removed it. I plan to replace it but haven't made time to do so. By the way, I really appreciate you watching and going through the videos with a fine tooth comb.

  • @joshdavis5224
    @joshdavis5224 ปีที่แล้ว

    At 24:06 should the g''(mu_i) term in the denominator actually be g'(mu_i). (First derivative as opposed to the second)

    • @statisticsmatt
      @statisticsmatt  ปีที่แล้ว +1

      Many thanks for bringing this to my attention! Much appreciated. You're absolutely correct. Many thanks for watching!

  • @changebio6417
    @changebio6417 3 ปีที่แล้ว +1

    Thanks for these helpful videos. I've only seen 3 videos before this one. I am wondering where is the fourth video?

    • @statisticsmatt
      @statisticsmatt  3 ปีที่แล้ว

      I think that you're the first to notice this. LOL. And, sorry. If my memory is correct, I think I combined video 4 with this video 5 to make them flow more coherently. It created a much longer video buy I think that it works better this way. If you have any questions, please let me know. And, many thanks for subscribing! Much appreciated.

  • @riskamulyani4801
    @riskamulyani4801 8 หลายเดือนก่อน

    hi,, in last min around 30:04, you did cancel both g'(ui), however in denumerator has g'(ui)^2 while numerator only g'(ui)...

    • @statisticsmatt
      @statisticsmatt  8 หลายเดือนก่อน

      Good question. here is the basis equation that you are asking about. 0=(1/g'(ui))*(...). multiple g'(ui) to the left side. 0*g'(ui) = 0.
      Many thanks for watching. Don't forget to subscribe and let others know about this channel.

  • @Songvbm
    @Songvbm 3 ปีที่แล้ว +1

    Matt, how irls is connected to maximum likelihood estimation?

    • @statisticsmatt
      @statisticsmatt  3 ปีที่แล้ว +2

      McCullagh and Nelder (1989) prove that the irls algorithm is equivalent to Fisher scoring and leads to maximum likelihood estimates.

    • @Songvbm
      @Songvbm 3 ปีที่แล้ว +1

      @@statisticsmatt Thanks. I need to check that. Also great content as usual.