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

    6
    ReputationRep:
    (Original post by paradoxequation)
    Has anyone done this question? In part a, you get 2.54 as an unbiased estimate for the variance, but then it straight up says that the variance is 2.56 despite this information only coming up in part b which hasn't happened yet.

    Also, in part b why do we use 2.56 and not 2.54? Thanks.
    DW about that they make hella printing errors.

    For part b we are given the official sd which we always choose over an estimated value
    Offline

    8
    ReputationRep:
    (Original post by paradoxequation)
    Has anyone done this question? In part a, you get 2.54 as an unbiased estimate for the variance, but then it straight up says that the variance is 2.56 despite this information only coming up in part b which hasn't happened yet.

    Also, in part b why do we use 2.56 and not 2.54? Thanks.
    Edit: There's no error. s^{2} is the unbiased estimator for variance, not \sigma^{2}
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by paradoxequation)
    Has anyone done this question? In part a, you get 2.54 as an unbiased estimate for the variance, but then it straight up says that the variance is 2.56 despite this information only coming up in part b which hasn't happened yet.

    Also, in part b why do we use 2.56 and not 2.54? Thanks.
    I don't quite understand what you are saying there.

    I suppose you could use a t test in part b but it has told you what the variance is.. so I guess that is the indicator that they want you to use the normal distribution.
    Offline

    6
    ReputationRep:
    (Original post by Euclidean)
    Edit: There's no error. s^{2} is the unbiased estimator for variance, not \sigma^{2}
    No there is they wrote 2.56 above the letter b which makes it look like an answer for part a when they actually write it for use in part b. I've seen this kinda thing before by them
    Offline

    8
    ReputationRep:
    (Original post by Rkai01)
    No there is they wrote 2.56 above the letter b which makes it look like an answer for part a when they actually write it for use in part b. I've seen this kinda thing before by them

    It's just badly presented. The \sigma^{2} is supposed to be next to (b) like you say. But the answer is there in the red box
    • Community Assistant
    Offline

    18
    ReputationRep:
    Community Assistant
    WOOOOOOWWWW the june 15 ial paper was a pile of piss in comparison to the june 15 normal paper what is up with that?!?!
    Offline

    8
    ReputationRep:
    (Original post by tazza ma razza)
    WOOOOOOWWWW the june 15 ial paper was a pile of piss in comparison to the june 15 normal paper what is up with that?!?!
    On that note, which past papers are most difficult for S3?

    I need to plan which ones I'll do in timed conditions as mocks, June 2015 for certain

    Tagging Zacken because he has a very good memory
    Offline

    18
    ReputationRep:
    do less degrees of freedom imply that the model is improved?
    Offline

    2
    ReputationRep:
    textbook says 'The continuous uniform distribution only has the requirement that the expected frequencies must equal the observed frequencies. Thus v = number of cells -1.'

    does this mean that u dont have to combine expected frequencies that are less than 5?
    • Community Assistant
    Offline

    18
    ReputationRep:
    Community Assistant
    (Original post by physicsmaths)
    do less degrees of freedom imply that the model is improved?
    uhhh yh i think so - check the graph of chi squared
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by connorbarr)
    textbook says 'The continuous uniform distribution only has the requirement that the expected frequencies must equal the observed frequencies. Thus v = number of cells -1.'

    does this mean that u dont have to combine expected frequencies that are less than 5?
    You would, then the nmber of cells would decrease accordingly (then take away one at the end).
    Offline

    2
    ReputationRep:
    (Original post by SeanFM)
    You would, then the nmber of cells would decrease accordingly (then take away one at the end).
    doesnt say anything about combining expected frequencies was why i asked, so should i always combine Ei < 5 regardless of the distribution?
    Offline

    6
    ReputationRep:
    (Original post by tazza ma razza)
    WOOOOOOWWWW the june 15 ial paper was a pile of piss in comparison to the june 15 normal paper what is up with that?!?!
    What you have done if q 4a actually asked "Find the probability that THREE randomly chosen sacks differ in weight by more than 0.5kg" e.g one weights 1kg another 1.6 and 2.3, how would you set the question up?
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by connorbarr)
    doesnt say anything about combining expected frequencies was why i asked, so should i always combine Ei < 5 regardless of the distribution?
    I would say so. It's unlikely that you would need to for the uniform distribution (there is a past paper question somewhere with a unif distribution) unless the widths are disproportionate - eg if you thought something was ~U(0,10) and one class was 0-0.5 and the next one was 0.5-3.
    • Community Assistant
    Offline

    18
    ReputationRep:
    Community Assistant
    (Original post by Rkai01)
    What you have done if q 4a actually asked "Find the probability that THREE randomly chosen sacks differ in weight by more than 0.5kg" e.g one weights 1kg another 1.6 and 2.3, how would you set the question up?
    hmmm

    i reckon i would @Zacken for help haha
    Offline

    2
    ReputationRep:
    (Original post by SeanFM)
    I would say so. It's unlikely that you would need to for the uniform distribution (there is a past paper question somewhere with a unif distribution) unless the widths are disproportionate - eg if you thought something was ~U(0,10) and one class was 0-0.5 and the next one was 0.5-3.
    sound thanks
    • Very Important Poster
    • Thread Starter
    Offline

    21
    ReputationRep:
    Very Important Poster
    (Original post by Rkai01)
    What you have done if q 4a actually asked "Find the probability that THREE randomly chosen sacks differ in weight by more than 0.5kg" e.g one weights 1kg another 1.6 and 2.3, how would you set the question up?
    I think that would just be a can of worms.

    The question is slightly ambiguous as it is - what does it mean for THREE things to differ by more than some weight?
    Offline

    3
    ReputationRep:
    (Original post by Euclidean)
    On that note, which past papers are most difficult for S3?

    I need to plan which ones I'll do in timed conditions as mocks, June 2015 for certain

    Tagging Zacken because he has a very good memory
    June 13R is a bit weirder but tbh they're all okay. Any hard solomon papers?
    Offline

    8
    ReputationRep:
    (Original post by Nikhilm)
    June 13R is a bit weirder but tbh they're all okay. Any hard solomon papers?
    They're all pretty standard

    I think S3 is distributed relatively uniformly in terms of difficulty.

    Feel free to test that hypothesis if you'd like!
    Offline

    11
    ReputationRep:
    (Original post by tazza ma razza)
    let me know how you do - i am doing it atm
    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?
 
 
 
  • 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
    Has a teacher ever helped you cheat?
  • 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

    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.