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

    11
    ReputationRep:
    (Original post by ChrisP97)
    I got something similar, I initially rounded both the first and last expected values up but was 0.1 over, so just rounded the last one (can't remember if that was 0.405) down.
    I think it was 12.005 for the first one, and 0.405 for the last one, so if you rounded to 12.01 and 0.41 you ended up being 0.01 over, so last one had to be adjusted to 0.40.
    Offline

    3
    ReputationRep:
    For the two assumptions make at the end of Q5 do you think 'That both data sets are assumed to be independant' is valid ?
    Thanks
    Offline

    3
    ReputationRep:
    (Original post by Unravelling)
    Did anyone else use the poisson formula to find the mean? I.e. e^-lambda = 16.53/100, and solve for lambda? I got the mean to be 1.7999... which gave me the same expected values and overall test statistic.
    I did that and got 1.799.

    I also worked out the mean from the table ( Sum of fx / sum of f ) and got 1.8, so i used 1.8.

    Either way, the value of lambda will give similar results
    Offline

    3
    ReputationRep:
    (Original post by econam)
    For the two assumptions make at the end of Q5 do you think 'That both data sets are assumed to be independant' is valid ?
    Thanks
    Yup. I think it'll be: both populations are independent and s^2 = sigma^2 for both populations
    Offline

    1
    ReputationRep:
    (Original post by AmarPatel98)
    I did that and got 1.799.

    I also worked out the mean from the table ( Sum of fx / sum of f ) and got 1.8, so i used 1.8.

    Either way, the value of lambda will give similar results
    I would assume both methods are allowed?
    Offline

    1
    ReputationRep:
    For the final question I got >43.xx
    Will I still get marks ?
    Offline

    3
    ReputationRep:
    (Original post by AmarPatel98)
    Yup. I think it'll be: both populations are independent and s^2 = sigma^2 for both populations
    I hope so because they are the ones I put! I have heard other people mention central limit theorem - maybe there is range of answers?
    Offline

    1
    ReputationRep:
    (Original post by adris)
    For the final question I got >43.xx
    Will I still get marks ?
    You might lose an answer mark, but I think that's it
    Offline

    0
    ReputationRep:
    [QUOTE=JoshC98;65165563]I think it was 12.005 for the first one, and 0.405 for the last one, so if you rounded to 12.01 and 0.41 you ended up being 0.01 over, so last one had to be adjusted to 0.40.[/QUO


    But I think rounding both of them would not make any differences to the answer. Must we just round either of them?!
    Offline

    1
    ReputationRep:
    The hypothesis killed me D-:
    Offline

    1
    ReputationRep:
    Btw everyone all hail king @arsey he is no longer with us. Edexcel have claimed him. There goes anyone chances of checking marks /-:
    Offline

    0
    ReputationRep:
    How did people do 4a??
    Offline

    3
    ReputationRep:
    (Original post by Unravelling)
    I would assume both methods are allowed?
    Perhaps, since some Ei values were given, whereas in past papers they usually just tell you to work out the mean in the first part of the question. Ultimately, we have to find lambda so i think it's likely they'll credit both methods.

    (Original post by econam)
    I hope so because they are the ones I put! I have heard other people mention central limit theorem - maybe there is range of answers?
    I dont think 'CLT applies' is a valid assumption. Usually, we mention that by the CLT, we assume the two sample means are normally distributed - but this answer is for the question 'What is the relevance of the CLT?'
    Offline

    3
    ReputationRep:
    Is there any way you can look back through your calculator memory (fx-991es)
    Offline

    0
    ReputationRep:
    (Original post by Fbiemad)
    Can someone else also confirm how many marks Q4 part (a) was?
    I heard few different answers (about the amount of marks for this question) so I dont know which one it is.

    Part (a) was about difference in E being greater than 2.
    P(I e1-e2 I > 2)
    What is your answer? Can you remember?
    Offline

    3
    ReputationRep:
    What is the edexcel policy for marking in the following case:

    The question was about the assumptions in the hypothesis test in our paper today. I wrote:
    'CLT applied so we assume the sample means are distributed normally.
    Assume the populations are independent.
    Assume s^2 = sigma^2'

    I think the last two are correct, but having written these three, will i get all the marks?
    Offline

    0
    ReputationRep:
    (Original post by Unravelling)
    I had no idea of any other way to do it :P When i asked other classmates, none of them did it our way and just calculated it from the tables (0 x its probability, 1 x its probability etc). which I didn't get since there were missing values we had to solve
    But I think rounding both of them would not make any differences to the answer. Must we just round either of them?!
    Offline

    0
    ReputationRep:
    (Original post by magixmike)
    How did people do 4a??
    Is the one about eggs?
    Offline

    0
    ReputationRep:
    (Original post by AmarPatel98)
    Perhaps, since some Ei values were given, whereas in past papers they usually just tell you to work out the mean in the first part of the question. Ultimately, we have to find lambda so i think it's likely they'll credit both methods.



    I dont think 'CLT applies' is a valid assumption. Usually, we mention that by the CLT, we assume the two sample means are normally distributed - but this answer is for the question 'What is the relevance of the CLT?'
    I think CLT is compulsory.
    Offline

    0
    ReputationRep:
    (Original post by AmarPatel98)
    What is the edexcel policy for marking in the following case:

    The question was about the assumptions in the hypothesis test in our paper today. I wrote:
    'CLT applied so we assume the sample means are distributed normally.
    Assume the populations are independent.
    Assume s^2 = sigma^2'

    I think the last two are correct, but having written these three, will i get all the marks?
    I wrote like this 1. S squared= sigma squared the sample variance is equal to the population variance. 2. The central limit theorem enables us to assume both of the sample means are normal. Because we are not told that the populations are normal but the sample sizes are large enough.

    Actually I think the first and the last one are correct. In terms of the past paper, every time there are this kind of questions
 
 
 
  • 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
    Brexit voters: Do you stand by your vote?
  • 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.