Hypothesis Testing (Binomial Distribution proportion) - Example 1 : ExamSolutions

แชร์
ฝัง
  • เผยแพร่เมื่อ 16 พ.ย. 2024

ความคิดเห็น • 40

  • @ladykoko2755
    @ladykoko2755 2 ปีที่แล้ว +9

    12 years now and students are still learning from you. Thank you!

  • @oliviaoneillx
    @oliviaoneillx 5 ปีที่แล้ว +16

    9 years old video and still helpful, thank you!

  • @Waddels
    @Waddels 5 หลายเดือนก่อน

    Thank you Max0r for introducing me to this absolute BANGER

  • @jiggasnap
    @jiggasnap 3 หลายเดือนก่อน

    I was always taught by my stats professors that you don't accept any hypothesis. You fail to reject. Great video though.

  • @ExamSolutions_Maths
    @ExamSolutions_Maths  12 ปีที่แล้ว +3

    Thanks. Good to hear that it helped.

    • @elliskase2845
      @elliskase2845 3 ปีที่แล้ว

      Not sure if anyone gives a damn but if you are stoned like me atm then you can stream pretty much all the latest movies on InstaFlixxer. I've been binge watching with my gf for the last few months xD

    • @kaysenemanuel4784
      @kaysenemanuel4784 3 ปีที่แล้ว

      @Ellis Kase Definitely, I've been using InstaFlixxer for months myself =)

  • @sarask2012
    @sarask2012 7 ปีที่แล้ว +18

    You do not "accept" H_0. You "do not reject".

    • @fifaheskey007
      @fifaheskey007 7 ปีที่แล้ว +9

      "insufficent evidence to reject h_0"

  • @ΝίκοςΖιώγος
    @ΝίκοςΖιώγος ปีที่แล้ว

    I've been taught that not rejecting Ho is not the same as accepting it,there just isn't enough evidence to reject Ho.There certainly isn't enough evidence to suggest the coin is fair,we just can't be 95% sure it isn't.(we can reject Ho with a slightly larger a)

  • @musicislife665
    @musicislife665 5 ปีที่แล้ว +1

    Thanks Professor! It was clear like always.

  • @leenav-cv1zf
    @leenav-cv1zf 7 หลายเดือนก่อน

    thanks but a doubt is its this situation a two tail test ?

  • @peaceandloveonthisplanetea6564
    @peaceandloveonthisplanetea6564 ปีที่แล้ว

    Super helpful 👍

  • @massimobuonaiuto8753
    @massimobuonaiuto8753 2 ปีที่แล้ว

    Great video. Super thank you

  • @lukeforryan
    @lukeforryan 12 ปีที่แล้ว +5

    Why do you need to do less than or equal to 1? I can't find anywhere that explains this. Why don't you just find out P(X=1)?

    • @yulindai5625
      @yulindai5625 6 ปีที่แล้ว

      I believe we should only count P(x=1), since the question is not no more than one time in 24 throws

    • @raphaelvannson3194
      @raphaelvannson3194 6 ปีที่แล้ว +4

      The P value looks at the probability of observing the results of the experiment *or a more extreme result* given that the null hypothesis is true.
      More extreme means "in the direction of the alternative hypothesis".
      I have asked myself this question as well, I do not have a really good principle / fundamental answer but here is how I live with this:
      1) This how how hypothesis tests are designed.
      2) It makes more sense when calculating p-values on continuous distributions since P(X = a number | Ho) = 0 when X is continuous. In this context it makes sense to ask for:
      P(X < a | Ho) (left-one-tailed, Ho: X=a, H1: X b | Ho) (right-one-tailed, Ho: X=a, H1: X>a)

    • @tamircohen1512
      @tamircohen1512 6 ปีที่แล้ว +5

      Only 6 years late but it is intuitive, no? Rolling 1 six in 24 throws it pretty unlikely, but the chance of rolling 1 or less sixes (i.e. rolling 1 six or 0 sixes) is actually greater than the chance of rolling just 1 six because these are 2 of the 25 possible outcomes rather than just one. However, rolling 1 six or 0 sixes is *more* extreme than rolling just one six and it gives more evidence that the die is biased.

    • @archjanasivasuthan6252
      @archjanasivasuthan6252 4 ปีที่แล้ว

      @@tamircohen1512 Thank you!!! Also why does the test statistic have to be less than the significance value?

    • @taaseen1941
      @taaseen1941 4 ปีที่แล้ว +3

      ​@@archjanasivasuthan6252 bit late to this but oh well. As you know, if the probability of getting a head on a supposedly fair coin is 1/2 and we have 24 trials, we would expect to get 12 heads. However, we could very well get much more than 12, like getting 22 heads or much less than 12, like 1, both of which are quite unlikely, causing you to say that it might be unfair. The question is, at what point do we say this is in fact not a fair coin? that's what the significance level is for. if we have a test statistic that has a probability of occurring which is less than the significance level given, we deem it "too" unlikely to support the null hypothesis. Bit of a lengthy explanation but hope that helps.

  • @tanya-uj4yq
    @tanya-uj4yq 4 ปีที่แล้ว +1

    thank you!

  • @ExamSolutions_Maths
    @ExamSolutions_Maths  14 ปีที่แล้ว

    Thanks

  • @th2
    @th2 14 ปีที่แล้ว

    thanks again

  • @ExamSolutions_Maths
    @ExamSolutions_Maths  12 ปีที่แล้ว

    @HanKooR Thanks

  • @ExamSolutions_Maths
    @ExamSolutions_Maths  12 ปีที่แล้ว

    What book are you using?

  • @MosesTan168
    @MosesTan168 12 ปีที่แล้ว

    Thank you for the help :)

  • @curryking33
    @curryking33 12 ปีที่แล้ว +2

    Why can't you just find the probability of X=1 and see if that is below the significance level, rather than doing the probability that X is less than or equal to 1? Thanks

  • @MosesTan168
    @MosesTan168 12 ปีที่แล้ว +2

    LOL THANK YOU AGAIN

  • @idontknowwhathandle2use
    @idontknowwhathandle2use 3 ปีที่แล้ว

    Please make a gaming channel

  • @ExamSolutions_Maths
    @ExamSolutions_Maths  14 ปีที่แล้ว

    @hongabonga12345 Cool

  • @TomerBenDavid
    @TomerBenDavid 4 ปีที่แล้ว

    :)

  • @leenav-cv1zf
    @leenav-cv1zf 7 หลายเดือนก่อน

    thanks