i.i.d. observations explained

Поділитися
Вставка
  • Опубліковано 7 лют 2025
  • Independent identically distributed observations (iid). What does iid mean exactly, and what 's the point of it?

КОМЕНТАРІ • 23

  • @jonathantieken
    @jonathantieken 11 років тому +27

    The most useful part for me was at t=2:55 where you "spelled out" exactly what being identically distributed meant. Thanks

    • @diegomabrary4974
      @diegomabrary4974 6 років тому

      I think he should prepare it more adequately...

  • @lac2275
    @lac2275 11 років тому +1

    agree like some in the comments, that the most helpful part is at 2:55 where you clarify what identically distributed meant. I am in college right now, and this also hadn't jumped to me until I watched your video. It clearly makes a huge difference when further delving into sample averages and variances as to why the results are what they are....Thanks Mate!

  • @sketchyAnalogies
    @sketchyAnalogies 2 роки тому

    Thanks! Using this for SIMIO, a simulation program for... loads of things. Theme Parks, factories, stadiums, hospitals, anything really. Here's me saying SIMIO more so people googling SIMIO get directed to here for SIMIO help.

  • @jayjayf9699
    @jayjayf9699 5 років тому +1

    Sir u make it very clear I don’t know why they over complicated it with rigorous notation that makes no sense to learners

  • @amizan8653
    @amizan8653 10 років тому +2

    Wow that cleared things up, thanks for the upload!

  • @perrygogas
    @perrygogas 9 років тому +1

    GREAT!!! the point in independence!!! I propose you make a new one based on this. Short and to the point.

  • @DuyNguyen-lo8xx
    @DuyNguyen-lo8xx 5 років тому

    Thank you, your video is really useful to me

  • @nicktanner8231
    @nicktanner8231 4 роки тому

    If i have a hat of 10 names and select three and after each selection I take it out, would that be considered identically distributed still or would I have to throw the pick back into the hat?

  • @Nats307
    @Nats307 2 роки тому

    I am confused with the i.i.d. observations vs i.i.d. variables. Are they the same thing or how can I "see" the difference?

  • @user93237
    @user93237 11 років тому

    Thanks! Great video.

  • @yongamer
    @yongamer 8 років тому +3

    2:24 I heared you said 好看. Thought it was worth mentioning.

    • @PhilChanstats
      @PhilChanstats  8 років тому

      Did you watch this just after chinese class? or watching chinese drama?! 谢谢

    • @yongamer
      @yongamer 8 років тому

      I tried to learn chinese for a couple years. Can't say I'm any good though :-P 你写汉语写的很好。

  • @shorndrack
    @shorndrack 12 років тому

    fantastic explanation..

  • @PhilChanstats
    @PhilChanstats  12 років тому

    Cheers

  • @tomvanbuuren9619
    @tomvanbuuren9619 5 років тому

    Thankyou

  • @mrzabalight
    @mrzabalight 8 років тому

    Is there a way to check if some observed variables are iid?

    • @PhilChanstats
      @PhilChanstats  8 років тому

      Hi. How you go about it depends on the nature of data (time series, cross section, panel). It's a short question that can not be answered in a few words.

    • @mrzabalight
      @mrzabalight 8 років тому +1

      The thing is i am working on a mobile nework simulator. The sources of
      the data are various and quiet complex. I found out some referebces that
      stated that we can use some tests such as Pearson correlation coeff or
      Kolmogorov Smirnoff test for Identical distributions and Run tests for
      idependency. However i think that all these tests are necessary, but not
      sufficient conditions to proove the iid property

    • @PhilChanstats
      @PhilChanstats  8 років тому +1

      i.i.d is a very strict assumption, but you must have reason. How you go about it depends on the nature of your data - survey? time series? Sounds like you are reading around. Read the methods that are relevant to the nature of your data - time series, cross section.