Hey there! Sign in to join this conversationNew here? Join for free
    • Community Assistant
    Online

    18
    ReputationRep:
    Community Assistant
    https://51625339b1524281e84f3037a08e...nVhdjA/CH3.pdf

    q9 ex 3A

    don't get part b)
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by tazza ma razza)
    https://51625339b1524281e84f3037a08e...nVhdjA/CH3.pdf

    q9 ex 3A

    don't get part b)
    How many ways are there of choosing 2 in 6?
    • Community Assistant
    Online

    18
    ReputationRep:
    Community Assistant
    (Original post by SeanFM)
    How many ways are there of choosing 2 in 6?
    yh but it mentions 7 draws
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by tazza ma razza)
    yh but it mentions 7 draws
    You're not wrong.. but that doesn't help for now.

    To get the 3rd one on the 7th draw, you don't care how you got them in the first 6 as long as you got 2 in the first 6, and then the 7th MUST be the 3rd one.

    Now think about my previous hint again.
    • Community Assistant
    Online

    18
    ReputationRep:
    Community Assistant
    (Original post by SeanFM)
    You're not wrong.. but that doesn't help for now.

    To get the 3rd one on the 7th draw, you don't care how you got them in the first 6 as long as you got 2 in the first 6, and then the 7th MUST be the 3rd one.

    Now think about my previous hint again.
    binomial 6c2 blah blah blah then

    kinda makes sense...
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by tazza ma razza)
    binomial 6c2 blah blah blah then

    kinda makes sense...
    If you are still stuck keep asking questions about it, or tell me which bits you can't see.
    • Community Assistant
    Online

    18
    ReputationRep:
    Community Assistant
    (Original post by SeanFM)
    If you are still stuck keep asking questions about it, or tell me which bits you can't see.
    haha dw as you can probably see, anything i don't get i post on the exam thread aha. Thanks btw, would rep but wont let me
    Offline

    1
    ReputationRep:
    When carrying out an F test for variability, do we need to include the lower F tail values as well? For example, S4 june 2012 q3 they show a calculation in brackets for the F test and the critical value in brackets, I don't understand where the critical value in the bracket comes from (I know we don't need to include the bracket answers).

    Lets say that question asked you to work out the lower critical value for the F test (assuming the test statistic was 225/36), the value shown in brackets is still wrong, I calculated 0.3257?
    Attachment 554205554209554240Attachment 554205554209554240
    Attached Images
      
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by A Question)
    When carrying out an F test for variability, do we need to include the lower F tail values as well? For example, S4 june 2012 q3 they show a calculation in brackets for the F test and the critical value in brackets, I don't understand where the critical value in the bracket comes from (I know we don't need to include the bracket answers).

    Lets say that question asked you to work out the lower critical value for the F test (assuming the test statistic was 225/36), the value shown in brackets is still wrong, I calculated 0.3257?
    Attachment 554205554209554240Attachment 554205554209554240
    I noticed that you have to as well. It is a two tailed test, after all. However, it feels like the mark scheme suggests that you can do the inverse (ie 36/225 instead of 225/36) and use the opposite tail for the test, which I suppose makes sense.

    How did you calculate 0.3257?
    Offline

    1
    ReputationRep:
    (Original post by SeanFM)
    I noticed that you have to as well. It is a two tailed test, after all. However, it feels like the mark scheme suggests that you can do the inverse (ie 36/225 instead of 225/36) and use the opposite tail for the test, which I suppose makes sense.

    How did you calculate 0.3257?
    The upper critical value at the 5% for F (10,8) = 3.35 from tables. Hence a lower critical value for that test would be 1/F(8,10). From tables, F (8,10) is 3.07 and so I calculated 0.3257. Can you verify this?

    It seems the mark scheme has not calculated lower critical values and has opted for an alternative method I guess. The only explanation that occurs to me is that a lower critical value can be ignored since if the F test statistic is close to one (zero) then it suggests that the variances are very close to each other and perhaps equal.
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by A Question)
    The upper critical value at the 5% for F (10,8) = 3.35 from tables. Hence a lower critical value for that test would be 1/F(8,10). From tables, F (8,10) is 3.07 and so I calculated 0.3257. Can you verify this?

    It seems the mark scheme has not calculated lower critical values and has opted for an alternative method I guess. The only explanation that occurs to me is that a lower critical value can be ignored since if the F test statistic is close to zero then it suggests that the variances are very close to each other and perhaps equal.
    I apologise - what they've calculated isn't the lower tail (where you would be right, it's 1/F(8,10) but they've done the test where, at every stage, they've done 1/test value, 1/critical value etc as a (rather strange) alternative method.

    I would like to say (although I am not sure how accurate it is), that, since they've got the test statistic and it is a rather large value, you know that you don't need to look at the lower tail so they don't bother with it.
    Offline

    1
    ReputationRep:
    (Original post by SeanFM)
    I apologise - what they've calculated isn't the lower tail (where you would be right, it's 1/F(8,10) but they've done the test where, at every stage, they've done 1/test value, 1/critical value etc as a (rather strange) alternative method.

    I would like to say (although I am not sure how accurate it is), that, since they've got the test statistic and it is a rather large value, you know that you don't need to look at the lower tail so they don't bother with it.
    Ok thanks. It appears that the convention is to use the larger sample variance to manipulate the test. The accuracy of the test I think is the same.

    Source: https://people.richland.edu/james/le...70/ch13-f.html
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by A Question)
    Ok thanks. It appears that the convention is to use the larger sample variance to manipulate the test. The accuracy of the test I think is the same.

    Source: https://people.richland.edu/james/le...70/ch13-f.html
    Ah I see, that explains it a lot better
    Offline

    8
    ReputationRep:
    When we calculate a critical region for a hypothesis test using a discrete distribution, is it always convention to go for an actual significance level which is less that the level they specify?

    In the question below they use < rather than ≈ which I normally use

    Spoiler:
    Show
    Offline

    6
    ReputationRep:
    How would you go about doing Q6a in the paper linked below:
    http://qualifications.pearson.com/co..._June_2006.pdf
    Offline

    3
    ReputationRep:
    I have 2 questions. 2013 R question 7 and 8. Especially 7
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by Euclidean)
    When we calculate a critical region for a hypothesis test using a discrete distribution, is it always convention to go for an actual significance level which is less that the level they specify?

    In the question below they use < rather than ≈ which I normally use
    Spoiler:
    Show
    The convention is that you go for the one that is closest to the significance level required. This could be above or below the significance level.
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by SargentZenj2)
    How would you go about doing Q6a in the paper linked below:
    http://qualifications.pearson.com/co..._June_2006.pdf
    What have you tried? it looks like you should use the definition of E(f(x)) (integration) to find it.
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by 130398)
    I have 2 questions. 2013 R question 7 and 8. Especially 7
    Please link the paper
    Offline

    8
    ReputationRep:
    (Original post by SeanFM)
    The convention is that you go for the one that is closest to the significance level required. This could be above or below the significance level.
    I thought so. I found it weird they'd used strictly below in that question. Thanks again!
 
 
 
  • See more of what you like on The Student Room

    You can personalise what you see on TSR. Tell us a little about yourself to get started.

  • Poll
    Brussels sprouts
  • See more of what you like on The Student Room

    You can personalise what you see on TSR. Tell us a little about yourself to get started.

  • The Student Room, Get Revising and Marked by Teachers are trading names of The Student Room Group Ltd.

    Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: International House, Queens Road, Brighton, BN1 3XE

    Quick reply
    Reputation gems: You get these gems as you gain rep from other members for making good contributions and giving helpful advice.