Hey there! Sign in to join this conversationNew here? Join for free
x Turn on thread page Beta
    Offline

    8
    ReputationRep:
    (Original post by econam)
    Im also confused by this. I thought because U1 U2... were all independant of each other then you don't square the constant of variation? Why is it not just over 5?
    Var(4U_1) = 16Var(U_1)

    Var(\frac{U_i}{5}) = \frac{1}{25}Var(U_i)

    Is it the movement from the whole Var function to individuals which has confused you?
    Online

    18
    ReputationRep:
    (Original post by SeanFM)
    Disagree what other exam do you get such lovely prose? except maybe those series questions in C1/C2.
    lol.
    M4-5 STEP and olympiads are enjoyable. step stats is enjoyable but s3 no way.
    this is how **** stats is, ch5 can be taught to an S1 candidate in 5 minutes.
    Offline

    8
    ReputationRep:
    (Original post by physicsmaths)
    i keep on applying the 0.5 **** since thats what the book did in examples.
    I did minimal questions from the book

    I think the S3 book is s***, no way around it

    (Original post by connorbarr)
    sound thanks, so i should do it for continuous uniform as well if there are gaps?
    Yup
    Offline

    2
    ReputationRep:
    (Original post by Euclidean)
    I did minimal questions from the book

    I think the S3 book is s***, no way around it



    Yup
    thanks a lot
    Online

    18
    ReputationRep:
    (Original post by Euclidean)
    I did minimal questions from the book

    I think the S3 book is s***, no way around it



    Yup
    so if i did a test of model suitable for normal and gaps were lets say
    5- 7 8-9 10-12 i wouldnt make them 7.5-9.5 etc?
    Offline

    3
    ReputationRep:
    s3 june 15 is 'fair' my ass
    Offline

    8
    ReputationRep:
    (Original post by physicsmaths)
    so if i did a test of model suitable for normal and gaps were lets say
    5- 7 8-9 10-12 i wouldnt make them 7.5-9.5 etc?
    edit: sorry I didn't see the 'wouldn't'

    You would apply a continuity correction because you're approximating something which is discrete (I presume?) by a distribution which is continuous

    physicsmaths
    Offline

    3
    ReputationRep:
    Is there an s3 paper from jan 16 IAL? The only IAL papers I can find are June 14 and June 15.

    Also, is there a June 13 Withdrawn?
    Offline

    8
    ReputationRep:
    (Original post by AmarPatel98)
    s3 june 15 is 'fair' my ass
    Did you get Var(U_1 - \bar{U})?

    (Original post by AmarPatel98)
    Is there an s3 paper from jan 16 IAL? The only IAL papers I can find are June 14 and June 15.Also, is there a June 13 Withdrawn?
    Don't think so
    Offline

    2
    ReputationRep:
    (Original post by physicsmaths)
    lol.
    M4-5 STEP and olympiads are enjoyable. step stats is enjoyable but s3 no way.
    this is how **** stats is, ch5 can be taught to an S1 candidate in 5 minutes.
    i did OCR maths so i did ch5 in S1
    Online

    18
    ReputationRep:
    (Original post by Euclidean)
    edit: sorry I didn't see the 'wouldn't'

    You would apply a continuity correction because you're approximating something which is discrete (I presume?) by a distribution which is continuous

    physicsmaths
    indeed so what were you saying you shouldn't do ?
    Offline

    9
    ReputationRep:
    Has anyone got a link for the June 15 IAL mark scheme? Cheers
    Offline

    3
    ReputationRep:
    (Original post by Zacken)
    Yep.
    You could 'assume' that the sample is large enough for the CLT to be accurate enough to implement?
    Offline

    3
    ReputationRep:
    (Original post by Euclidean)
    Did you get Var(U_1 - \bar{U})?



    Don't think so
    Hell no but ngl i thought it was a nice question, so if they throw it as us again, we're ready!
    Offline

    22
    ReputationRep:
    (Original post by Nikhilm)
    You could 'assume' that the sample is large enough for the CLT to be accurate enough to implement?
    No, if n \geq 50, then CLT is invoked. If the question says n = 70, then you don't "assume 70 > 50", do you?

    (Original post by physicsmaths)
    so if i did a test of model suitable for normal and gaps were lets say
    5- 7 8-9 10-12 i wouldnt make them 7.5-9.5 etc?
    (Original post by Euclidean)
    edit: sorry I didn't see the 'wouldn't'

    You would apply a continuity correction because you're approximating something which is discrete (I presume?) by a distribution which is continuous

    physicsmaths
    You would make them 7.5 - 9.5.

    (Original post by AmarPatel98)
    s3 june 15 is 'fair' my ass
    It is. What did you dislike about it?
    Offline

    22
    ReputationRep:
    (Original post by ChrisP97)
    Has anyone got a link for the June 15 IAL mark scheme? Cheers
    It's on the Edexcel website, just navigate through: our quals -> internationa a-levels -> mathematics -> exam materials -> S3
    Offline

    19
    ReputationRep:
    (Original post by AmarPatel98)
    s3 june 15 is 'fair' my ass
    What's unfair about it?
    Offline

    3
    ReputationRep:
    (Original post by Euclidean)
    Var(4U_1) = 16Var(U_1)

    Var(\frac{U_i}{5}) = \frac{1}{25}Var(U_i)

    Is it the movement from the whole Var function to individuals which has confused you?
    Had a look at the mark scheme - I feel I understand it now. Combining 4U1 the get 16sigma
    Offline

    3
    ReputationRep:
    Its actually this for normal distribution chi squared tests if you refer to the book.
    Attached Images
     
    Offline

    22
    ReputationRep:
    (Original post by kennz)
    Its actually this for normal distribution chi squared tests if you refer to the book.
    If it's a normal distribution then the greater than or equal to or w/e doesn't make a difference.
 
 
 
Reply
Submit reply
Turn on thread page Beta
Updated: June 29, 2016
Poll
Do you agree with the proposed ban on plastic straws and cotton buds?

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

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