Statistics 101: Nonlinear Regression, The Quadratic Model

Поділитися
Вставка
  • Опубліковано 18 січ 2025

КОМЕНТАРІ • 26

  • @mohammedalsubaie3512
    @mohammedalsubaie3512 3 роки тому

    I was that quite student in the class without any participation, but now! Oh man I even stand up in front of the class and explain what everything comes together. Thank you from the bottom of my heart

  • @pavittarsingh5870
    @pavittarsingh5870 6 років тому +5

    I am happy the channel is still active, and videos are being uploaded. I hope you keep it up... The tutorials are great, really love them. Tbh, I have tried a few out there but these were the awesomest from all I can see. :)
    In further, I would request you to, please upload the slides (pdf would be fine), if possible. Its hard to make notes at the same time and enjoy the video at the same time, I feel like I am going to miss on something.. Thanks.

  • @bensmith6987
    @bensmith6987 5 років тому +3

    This is much better quality both in delivery the content and lay out. Thank you so much. Should have a course on udemy. Then i would have taken it 100%.

  • @swghavoc
    @swghavoc 2 роки тому

    Wow! Thank you for making it understandable!

  • @tantruongkhanh1312
    @tantruongkhanh1312 Рік тому

    Thank you PROF

  • @theos.6229
    @theos.6229 6 років тому +1

    amazing like always...your videos are really made to facilitate understanding...Great Job Sir!

  • @fredygonzalocarvajalserran1654
    @fredygonzalocarvajalserran1654 6 років тому +1

    Nice video, thanks...Excellent explanation.

  • @nawilliam2754
    @nawilliam2754 4 роки тому

    Great video. Indebted to you Brandon.

  • @pierre-louispaugam5773
    @pierre-louispaugam5773 6 років тому +3

    Hello Brandon I have one question.
    Do you plan on making an "ANCOVA - Understanding the calculations" video one day ? :)
    Thank you in advance !

  • @coscinaippogrifo
    @coscinaippogrifo 4 роки тому

    Bless you for these amazing videos! You're making my quarantine enjoyable and useful :)

  • @sintiacarolina1122
    @sintiacarolina1122 4 роки тому

    really easy to understand. thank you, it helps me alot. :)

  • @subratasutradhar3599
    @subratasutradhar3599 Рік тому

    Which software did you use to do the calculation?

  • @edgarbahilorodriguez3066
    @edgarbahilorodriguez3066 6 років тому +3

    Hi! I am sorry but using a predictor^2 does not change your OLS problem to a non-linear one.
    The vector of least-squares estimators a1,b1,c1, ... depends in a linear way on the vector of response variables y1,y2,y3,…,yn. You can not say that a quadratic model is not linear because it can be expressed as a linear combination of the predictors.
    I think that you need to clarify that even that you are fitting a non-linear relationship between the inputs and the outputs from the statistical point of view the problem is still being linear.
    BR

  • @charlesschwer1792
    @charlesschwer1792 6 років тому

    Great video! In reality, due to job turn over, there would not be as many salesmen still working after five years compared to the number of salesmen who have worked less than three years. So the density of observations should show more than half who had 3.5 years or less experience. Unless, there is almost no job turn over.

  • @karnam7444
    @karnam7444 5 років тому +1

    Hi Brandon, do you know how I can manually calculate sum of square of quadratic factors such as A^2, B^2 etc?

  • @JuanCarloshgo
    @JuanCarloshgo 6 років тому

    hey, I love your videos, do you plan to do something on bayesian statistics??

  • @tnl6349
    @tnl6349 4 роки тому

    Hi Brandon
    Thanks for this great video.
    I was wondering if it was possible and if yes how would look the regression equation for a quadratic model with polynomial x with multiple independant variable?
    Thanks

  • @KhaLed-pb4pu
    @KhaLed-pb4pu 5 років тому

    how did you decide that b1 is the x term and b2 is the xsquared term?

  • @lauAlv
    @lauAlv 5 років тому +2

    "now, don't freak out, this is not that bad". HOW DO YOU KNOW US SO WELL? lol.

  • @bobjohnson4318
    @bobjohnson4318 6 років тому

    Hi Brandon, a request. I am interested in understanding the relationship of Design of Experiments, like a Central Composite Design, with Regression. Are they different subjects, or is Regression a component of DOE? Also, I think I understand interaction effects, but quadratics --- I cannot grasp them yet. I really like how you slow down and explain what is going on, answering all the questions that a novice would ask themselves as they went through the process. We like understanding what the calculations are, even though we will ultimately let the software do it. Any thoughts on this are appreciated, or if you want to ask me for any clarification, I hope I am explaining it right.

  • @raymondgeorge1545
    @raymondgeorge1545 4 роки тому

    Thank you Brandon it was really helpful in my professional transformation.
    But When I checked the maximum point is 380.49..... so is it fine to take only few 4 number after decimal for accuracy 🤔?

  • @Gandhimathy-o4f
    @Gandhimathy-o4f Рік тому

    Super

  • @ccuuttww
    @ccuuttww 6 років тому

    sometimes u may find multi min if your curve have more argument
    Its can a same thing in gradient descent

    • @BrandonFoltz
      @BrandonFoltz  6 років тому

      Absolutely! This was a simple example. I will get to gradient descent here soon. Thanks Jill!

  • @ashrukhtf3183
    @ashrukhtf3183 5 років тому

    Can you please share one example with python for non linear regression example

  • @muratcanakdogan26
    @muratcanakdogan26 4 роки тому

    Hello y=1/(Ax+B)^2 and y=A/(x+B) least square table fitting How to solve?