Normality Testing of Regression Errors | Testing OLS residuals for Normality

Поділитися
Вставка
  • Опубліковано 1 гру 2024

КОМЕНТАРІ • 18

  • @عطاءالرحمن
    @عطاءالرحمن 4 роки тому +1

    Great to see you back sir.

  • @abdulsubhan4026
    @abdulsubhan4026 4 роки тому +1

    Very Informative😍

    • @FAMFAMI
      @FAMFAMI  4 роки тому +1

      Thanks a bundle!!

  • @saniashaheenvisitingteache396
    @saniashaheenvisitingteache396 4 роки тому +1

    Amazing lecture sir

  • @amaraahmed927
    @amaraahmed927 4 роки тому +1

    Thank you for the sharing

  • @basitalikhattak7129
    @basitalikhattak7129 4 роки тому +1

    After a very long break sir. ❤️❤️❤️

    • @FAMFAMI
      @FAMFAMI  4 роки тому

      From now on I will be regular IA!

  • @WanderlensbyJawad
    @WanderlensbyJawad 4 роки тому +1

    Sir ❤️

  • @fatimaniazi454
    @fatimaniazi454 3 роки тому +1

    Murshad ❤️😁

  • @sukursukur3617
    @sukursukur3617 3 роки тому +1

    Why normality tests? Why dont we implement poissonity test? What makes normal distribution privileged among other distributions?

    • @FAMFAMI
      @FAMFAMI  3 роки тому

      Normality is imp to make valid inferences--especially in small samples--in large samples one can rely on Central limit theorem and can ignore normality

    • @sukursukur3617
      @sukursukur3617 3 роки тому

      @@FAMFAMI firstly thank you for answer. Can you say all fusses about normality tests stem from trying to infer meaning from small dataset?

    • @FAMFAMI
      @FAMFAMI  3 роки тому

      If u look at analytical formula for the OLS estimates of true parameters then u wil realize that these are linear combination of true errors. By assuming normality of errors the inference regarding significance may b different.
      Since true errors r unknown and unobserved so usually residuals obtained from OLS regression are tested for normality