Hey there! Sign in to join this conversationNew here? Join for free
    Offline

    2
    ReputationRep:
    Can someone explain why in Example 6 page 78, n=5 and not 6?
    Offline

    0
    ReputationRep:
    Hi, could somebody help me with this question, part b? It's from June 2009
    Name:  june 2009 s3 question.png
Views: 116
Size:  45.3 KB
    Offline

    6
    ReputationRep:
    (Original post by ddbrain)
    Hi, could somebody help me with this question, part b? It's from June 2009
    Name:  june 2009 s3 question.png
Views: 116
Size:  45.3 KB
    Don't know why I wrote xbar weren't needed
    Attached Images
     
    Offline

    0
    ReputationRep:
    (Original post by Rkai01)
    Don't know why I wrote xbar weren't needed
    Thank you so much
    Offline

    2
    ReputationRep:
    (Original post by gagafacea1)
    With contingency tables, what do we do if we get an expected frequency less than 5, I've never seen one, but what if?

    Also in the conclusions for those questions, do we usually say "not enough evidence for association" or "they are not independent", because I'm getting mixed messages from the mark schemes. Sometimes they seem to be sure like "Money has NO relation to happiness" and sometimes they're like "Oh, I guess we don't have enough evidence of a relationship between money and happiness". I'm like, *****, MAKE! UP! YOUR! MIND!!
    When you have Ei less than 5, and there are 3 of them say which add up to more than 5, do you add them up in one column or do you add it to the column to the left.
    Offline

    18
    ReputationRep:
    Gna start papers tmmrw night. Im guessing the only hard paper(generally) is june 15?
    Ill just do 2010 onwards.
    Goodness of fit tests must be around 10 marks atleast right?


    Posted from TSR Mobile
    Offline

    8
    ReputationRep:
    (Original post by physicsmaths)
    Gna start papers tmmrw night. Im guessing the only hard paper(generally) is june 15?
    Ill just do 2010 onwards.
    Goodness of fit tests must be around 10 marks atleast right?


    Posted from TSR Mobile
    I asked this earlier, I think 15, 15 IAL and 13 R are the difficult ones

    Goodness of fit are normally at least 10ish
    Offline

    18
    ReputationRep:
    (Original post by Euclidean)
    I asked this earlier, I think 15, 15 IAL and 13 R are the difficult ones

    Goodness of fit are normally at least 10ish
    decent. the s3 content is extremely easy. good i didnt spend more time on this module as it aint worth it lol.
    Offline

    0
    ReputationRep:
    Yo lads and lass's - can anyone explain why a parameter that is estimated by calculation, sacrifices a degree of freedom?

    I assume its because you loose a degree of freedom with the calculation I don't really fully understand it tho!
    Offline

    22
    ReputationRep:
    (Original post by physicsmaths)
    Ill just do 2010 onwards.e
    That's what I'm doing.
    Offline

    11
    ReputationRep:
    On the june 2015 paper:

    for q5 part aii, am I right in saying that E(D) = E(Y1) - E(xbar) = mu- mu = 0 ?

    if so why does E(Y1) = mu? Can someone explain what's going on here



    Zacken <3
    Attached Images
     
    Offline

    3
    ReputationRep:
    (Original post by Euclidean)
    I asked this earlier, I think 15, 15 IAL and 13 R are the difficult ones

    Goodness of fit are normally at least 10ish
    Agree with this but what was so difficult about the IAL? I thought it was standard
    Offline

    22
    ReputationRep:
    (Original post by Katiee224)
    On the june 2015 paper:

    for q5 part aii, am I right in saying that E(D) = E(Y1) - E(xbar) = mu- mu = 0 ?

    if so why does E(Y1) = mu? Can someone explain what's going on here



    Zacken <3
    Yep. Basically we're sampling from the normal population, so each sample has the same distribution as the population. i.e: Y_1 \sim N(\mu, \sigma^2) since we're sampling from N(\mu, \sigma^2).

    I get a whole heart, eh?
    • Community Assistant
    Offline

    18
    ReputationRep:
    Community Assistant
    (Original post by Katiee224)
    I just finished it and tbh it was the hardest s3 paper I've sat, I really struggled with the latter part of q5 but the rest was doable, how did you find it?
    look at post 606 - my thoughts on the IAL 2015 vs normal 2015 paper

    i got 70/75 (bang on 90 UMS according to the ums calculator) on the ial. i am resitting s3 - got 56 ums in the 2015 paper last year - that paper ruined me haha gonna try and get 85+ ums on it tomorrow morning

    june 2015 is without doubt the worst paper for s3, leaps and bounds more difficult than the next hardest tbph
    Offline

    2
    ReputationRep:
    Why is it that sometimes the formula for a z test is x - u / (standard deviation / root n) but other times it's just X - u / standard deviation? Can anyone explain?
    Offline

    17
    ReputationRep:
    (Original post by etmarks)
    Yo lads and lass's - can anyone explain why a parameter that is estimated by calculation, sacrifices a degree of freedom?

    I assume its because you loose a degree of freedom with the calculation I don't really fully understand it tho!
    Let's consider a set of, say, n bits of data as in a typical S3 question.

    If there were absolutely no restrictions on the values you choose, you would have n degrees of freedom, clearly - if you lose one piece of expected data you can't hope to find out what it was.

    However, in the vast majority of cases the sum of your expected values is fixed at the number of observations. For this, you lose a degree of freedom - you can now lose up to one piece of data and still recover it, since you know the total and you know all the other expected values.

    Now let's say you calculated a known mean based on youe initial observations. In this case we lose one more degree of freedom. Even if we lost two expected values, we could work out their sum as discussed above. However, since we know the average, we can now get another equation linking these two unknown values. Two equations, two unknowns - we can sort that out.

    This is a somewhat backward way of thinking about degrees of freedom, I suppose. But fundamentally, each new 'fact' about your expected values gets you another equation, which allows you to 'lose' one more piece of data and thus reduces the degrees of freedom by one.

    Posted from TSR Mobile
    Offline

    8
    ReputationRep:
    (Original post by Ayman!)
    Agree with this but what was so difficult about the IAL? I thought it was standard
    I'm not sure, I asked to see which papers I should do in timed proper conditions and someone mentioned 15 IAL. Do you know of any tough ones off the top of your head?

    (Original post by physicsmaths)
    decent. the s3 content is extremely easy. good i didnt spend more time on this module as it aint worth it lol.
    You wouldn't wanna spend any more time, I started same time as you after M3 and I'm bored already :lol:
    Offline

    18
    ReputationRep:
    (Original post by Euclidean)
    I'm not sure, I asked to see which papers I should do in timed proper conditions and someone mentioned 15 IAL. Do you know of any tough ones off the top of your head?



    You wouldn't wanna spend any more time, I started same time as you after M3 and I'm bored already :lol:
    Yeh its propa **** init.


    Posted from TSR Mobile
    Offline

    18
    ReputationRep:
    (Original post by coolguy123456)
    Why is it that sometimes the formula for a z test is x - u / (standard deviation / root n) but other times it's just X - u / standard deviation? Can anyone explain?
    1st one is approximate estimation i think. Second one is proper.


    Posted from TSR Mobile
    Offline

    2
    ReputationRep:
    how are the observed frequencies calculated in this question.
 
 
 
  • 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
    Would you like to hibernate through the winter months?
  • 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.