Turn on thread page Beta
    Offline

    3
    ReputationRep:
    (Original post by Zacken)
    Pretty sure the textbook says the proof isn't examinable.
    It is Name:  Q.jpg
Views: 78
Size:  17.2 KB June 2014 Question 6.

    I'm not sure why a proof like that would only be 1 mark?
    Online

    22
    ReputationRep:
    (Original post by Inges)
    Its not letting me upload it so i put it on imgur.

    Putting it up here for you, hope you don't mind:
    Offline

    3
    ReputationRep:
    (Original post by Inges)
    When you write Sum of O2 / E , You have to write every number your adding together, basically showing how your adding all your observed frequenices squared divided by your expected frequencies and how your taking away the original Total and then how you got the Chi-squared value. Then compared the Chi-squared value with the critical value from the table showing if its in the critical region or not. Then stating your conclusion.
    Thanks!

    What's this total error and total mean error i've seen people talking about?!
    Online

    22
    ReputationRep:
    (Original post by Inges)
    It is Name:  Q.jpg
Views: 78
Size:  17.2 KB June 2014 Question 6.

    I'm not sure why a proof like that would only be 1 mark?
    I thought he was asking about proving that s^2 was an unbiased estimator of \sigma^2 which isn't examinable. (un)biased estimators for other functions are most definitely examinable though, yes.
    Offline

    3
    ReputationRep:
    (Original post by Zacken)
    Putting it up here for you, hope you don't mind:
    Of course not, I don't know whats wrong with my browser freezing when I do it ;P
    Online

    22
    ReputationRep:
    (Original post by AmarPatel98)
    What's this total error and total mean error i've seen people talking about?!
    Nothing. Ignore it, it's coming from nowhere.
    Offline

    3
    ReputationRep:
    (Original post by AmarPatel98)
    Thanks!

    What's this total error and total mean error i've seen people talking about?!
    Anytime & were on the same page buddy I don't know what total error is to be frank!
    Offline

    3
    ReputationRep:
    (Original post by Zacken)
    I thought he was asking about proving that s^2 was an unbiased estimator of \sigma^2 which isn't examinable. (un)biased estimators for other functions are most definitely examinable though, yes.
    Oh right!
    Offline

    2
    ReputationRep:
    (Original post by Zacken)
    I thought he was asking about proving that s^2 was an unbiased estimator of \sigma^2 which isn't examinable. (un)biased estimators for other functions are most definitely examinable though, yes.
    Why isn't it examinable? It's a fairly straight forward proof...
    Offline

    3
    ReputationRep:
    (Original post by Zacken)
    Nothing. Ignore it, it's coming from nowhere.
    HAHA! I reckon its people doing IAL, Zacken is that right? ( I know nothing about IAL )
    Offline

    2
    ReputationRep:
    (Original post by Zacken)
    It's because U_1 and \bar{U} are not independent. The formula you quoted are only for independent random variables X,Y. They are not independent because they come from the same sample, but the important thing here is that they are not independent.
    Cool, I see that now. But then isn't the method to working out the variance as 4sigma^2/5 using the same formula?

    Is the mark scheme attached not using Var(F) = Var(U1-Ubar) = Var(U1) + Var(Ubar) ??
    What is different about it?
    Attached Images
     
    Online

    22
    ReputationRep:
    (Original post by Armpits)
    Why isn't it examinable? It's a fairly straight forward proof...
    It's fairly algebraically dense, see page 29 of the textbook. Anyhow, lots of proofs are straighforward, doesn't mean they're examinable.

    (Original post by Inges)
    HAHA! I reckon its people doing IAL, Zacken is that right? ( I know nothing about IAL )
    Nope, IAL and UK have the exact same specification. Total error doesn't mean anything in terms of S3-level stuff.
    Online

    22
    ReputationRep:
    (Original post by paradoxequation)
    Cool, I see that now. But then isn't the method to working out the variance as 4sigma^2/5 using the same formula?

    Is the mark scheme attached not using Var(F) = Var(U1-Ubar) = Var(U1) + Var(Ubar) ??
    What is different about it?
    It is using the same formula because it's written \bar{U} in terms of U_1, U_2, \cdots, U_5 all of which are independent of one another.
    Offline

    2
    ReputationRep:
    (Original post by Zacken)
    It is using the same formula because it's written \bar{U} in terms of U_1, U_2, \cdots, U_5 all of which are independent of one another.
    Thank you very much, it makes sense to me now.

    One last question, why are Ubar and U1 not independent to each other, but U1, U2, ..., U5 are?
    Offline

    3
    ReputationRep:
    (Original post by Zacken)
    It's fairly algebraically dense, see page 29 of the textbook. Anyhow, lots of proofs are straighforward, doesn't mean they're examinable.



    Nope, IAL and UK have the exact same specification. Total error doesn't mean anything in terms of S3-level stuff.
    Ohhh, someone really shook the whole forum up!
    Online

    18
    ReputationRep:
    regarding total error, it was a question in the textbook talking about error in some contextual problem not as in anything to do with standard error or anything.
    Online

    22
    ReputationRep:
    (Original post by paradoxequation)
    Thank you very much, it makes sense to me now.

    One last question, why are Ubar and U1 not independent to each other, but U1, U2, ..., U5 are?
    Well, think of it, two things being independent means knowing something about one thing gives you no information about the other. And the other way around for dependent.

    So, if we know something about U1, that gives us a bit of information about Ubar since we calculate Ubar using U1 (and other stuff), but the key thing is that knowing U1 or U2 or U3 or U4 or U5 gives us information about Ubar.

    So they are not independent.

    However, knowing something about U1 gives you no information whatsoever about U2 or U3 or anything, hence U1 and U2 are independent, same for U2 and U3 and U1 and U3 and etc...

    Remember that just because the things are all independent to one another (U1 independent to U2 to U3 to U4, etc...) doesn't mean that the sum of those things (Ubar) is also independent of them.

    [Lost the will to LaTeX...]

    (Original post by Inges)
    Ohhh, someone really shook the whole forum up!
    Yep! Love your handwriting, by the way! And thank you for helping out tonight, really saved me from having to answer tons of questions.
    Offline

    2
    ReputationRep:
    Spearman rank correlation coefficient, when ranking the values, if you rank biggest value of x, 1 and smallest 10, then you do the same for y even if it gives you a negative value which means a negative correlation?
    Offline

    2
    ReputationRep:
    (Original post by Zacken)
    Well, think of it, two things being independent means knowing something about one thing gives you no information about the other. And the other way around for dependent.

    So, if we know something about U1, that gives us a bit of information about Ubar since we calculate Ubar using U1 (and other stuff), but the key thing is that knowing U1 or U2 or U3 or U4 or U5 gives us information about Ubar.

    So they are not independent.

    However, knowing something about U1 gives you no information whatsoever about U2 or U3 or anything, hence U1 and U2 are independent, same for U2 and U3 and U1 and U3 and etc...

    Remember that just because the things are all independent to one another (U1 independent to U2 to U3 to U4, etc...) doesn't mean that the sum of those things (Ubar) is also independent of them.

    [Lost the will to LaTeX...]
    Haha, no problem! I fully understand the independence thing now, thank you. If only the textbook was as clear as your description lol.
    Offline

    2
    ReputationRep:
    just run through my note on significant testing and i wrote " for PMCC, variables have to follow normal distribution", and i have no idea what this mean and how this applies ..........
 
 
 
Reply
Submit reply
Turn on thread page Beta
Updated: June 29, 2016
The home of Results and Clearing

2,353

people online now

1,567,000

students helped last year

University open days

  1. Sheffield Hallam University
    City Campus Undergraduate
    Tue, 21 Aug '18
  2. Bournemouth University
    Clearing Open Day Undergraduate
    Wed, 22 Aug '18
  3. University of Buckingham
    Postgraduate Open Evening Postgraduate
    Thu, 23 Aug '18
Poll
A-level students - how do you feel about your results?

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.