Turn on thread page Beta
    • Thread Starter
    Offline

    3
    ReputationRep:
    (Original post by 000alex)
    What are the exact things I need to look for in a question to see if I have to use the CLT or not? I really don't get it >.<
    If they give you a distribution of the mean and n is large then you should probably use CLT

    Edit: or above
    Offline

    11
    ReputationRep:
    How's everyone feeling for tomorrow? I think I've finally cracked S2!
    • Thread Starter
    Offline

    3
    ReputationRep:
    (Original post by rich1334)
    How's everyone feeling for tomorrow? I think I've finally cracked S2!
    I feel like I'm waaay too relaxed for this exam :/ No idea why Good luck on S2 tomorrow as well!
    Offline

    11
    ReputationRep:
    (Original post by lllllllllll)
    I feel like I'm waaay too relaxed for this exam :/ No idea why Good luck on S2 tomorrow as well!
    Same, I feel like I'm going to be complacent and then be hit by some awful question in the exam Good luck to you too!
    Offline

    2
    ReputationRep:
    That's was horrible. Couldn't do question 4 for gods sake.


    Posted from TSR Mobile
    • Thread Starter
    Offline

    3
    ReputationRep:
    I may have gotten question 8(II) wrong I know it was a small number but were we meant to work out P(Type-II) using using the significance level or critical value?


    Posted from TSR Mobile
    • Thread Starter
    Offline

    3
    ReputationRep:
    (Original post by Sam596)
    That's was horrible. Couldn't do question 4 for gods sake.


    Posted from TSR Mobile
    Was that the one about Type-1/2 yprobabilities?

    Edit: just realised it is. Dw I don't think a lot of people got that

    Posted from TSR Mobile
    Offline

    13
    ReputationRep:
    I know I've screwed up question 8(ii) but *touchwood* the rest went okay.
    Offline

    2
    ReputationRep:
    Question 4 was two binomial approx, one poisson one normal right? iirc it was a normally distributed function so type 1 error probability is the significance level and type 2 was given, then you modelled it as a binomial function (it either has an error or doesnt) and do the standard for approximating

    that was a cute context from ocr on question 2 - dead bunnies along a road?

    also is anyone feeling confident in their 8ii answer? im not sure mine is right but it was the only thing i could think of (I'll type it up once im not on mobile)
    Offline

    11
    ReputationRep:
    (Original post by ms8061)
    Question 4 was two binomial approx, one poisson one normal right? iirc it was a normally distributed function so type 1 error probability is the significance level and type 2 was given, then you modelled it as a binomial function (it either has an error or doesnt) and do the standard for approximating

    that was a cute context from ocr on question 2 - dead bunnies along a road?

    also is anyone feeling confident in their 8ii answer? im not sure mine is right but it was the only thing i could think of (I'll type it up once im not on mobile)
    I got poisson and then normal approximations for Q4 too.

    Very last question I did

    (1/3*P(X<8|p=0.5) + 1/3*P(X<8|p=0.7) )* 4 and got 0.931 or something like that
    Offline

    0
    ReputationRep:
    LIKE WHAT WAS UP WITH THAT EXAM!!!
    Question 4, couldnt even attempt!!!
    Offline

    1
    ReputationRep:
    I felt pretty good about all my answers, ill post as many as i can remember.

    a=6 then a=1.5 for cont dist qn

    Mean = 146, S.D=2 for normal qn

    Very last qn was something like:
    1/3(0.6 + 0.03) <-- i really cant remember
    So P(type 2 error) = 0.23 ish

    So P(X at least 1 out of 4) = 1 - P(X=0) = 1- (1-0.23)^4
    Offline

    13
    ReputationRep:
    (Original post by ms8061)
    Question 4 was two binomial approx, one poisson one normal right? iirc it was a normally distributed function so type 1 error probability is the significance level and type 2 was given, then you modelled it as a binomial function (it either has an error or doesnt) and do the standard for approximating

    that was a cute context from ocr on question 2 - dead bunnies along a road?

    also is anyone feeling confident in their 8ii answer? im not sure mine is right but it was the only thing i could think of (I'll type it up once im not on mobile)
    (Original post by rich1334)
    I got poisson and then normal approximations for Q4 too.

    Very last question I did

    (1/3*P(X<8|p=0.5) + 1/3*P(X<8|p=0.7) )* 4 and got 0.931 or something like that
    Yup, I got poisson and normal distributions too.

    I reckon I've got 2 marks for 8(ii) as I worked out the P(Type II) at 0.5 and 0.7 but what I've done after that was just clutching at straws. My friend made it a binomial distribution?
    Offline

    11
    ReputationRep:
    I'm pretty sure my method is partially wrong, could very much be a binomial

    (Original post by Molly_xox)
    Yup, I got poisson and normal distributions too.

    I reckon I've got 2 marks for 8(ii) as I worked out the P(Type II) at 0.5 and 0.7 but what I've done after that was just clutching at straws. My friend made it a binomial distribution?
    Offline

    2
    ReputationRep:
    Thank god I'm not the only one who found Q4 impossible.


    Posted from TSR Mobile
    Offline

    1
    ReputationRep:
    (Original post by Molly_xox)
    Yup, I got poisson and normal distributions too.

    I reckon I've got 2 marks for 8(ii) as I worked out the P(Type II) at 0.5 and 0.7 but what I've done after that was just clutching at straws. My friend made it a binomial distribution?
    once you found P(type 2 error) about 0.23 i think . you had to find the probability of getting at least 1 out of 4 tests. so its a binomial B(4 , 0.23)
    Offline

    6
    ReputationRep:
    Did question 1 ask for standard deviation or variance? Got sigma = 2 but don't wanna lose a cheap mark for not stating variance =4...
    Offline

    11
    ReputationRep:
    Yeah that looks good. I stupidly multiplied my P(Type II) by 4 instead of treating it like binomial. Oh well, probably got )2/3 marks for it

    (Original post by tokopoko)
    I felt pretty good about all my answers, ill post as many as i can remember.

    a=6 then a=1.5 for cont dist qn

    Mean = 146, Var =4 for normal qn

    Very last qn was something like:
    1/3(0.6 + 0.03) <-- i really cant remember
    So P(type 2 error) = 0.23 ish

    So P(X at least 1 out of 4) = 1 - P(X=0) = 1- (1-0.23)^4
    Offline

    13
    ReputationRep:
    (Original post by Tomh97)
    Did question 1 ask for standard deviation or variance? Got sigma = 2 but don't wanna lose a cheap mark for not stating variance =4...
    I thought it asked for standard deviation too!
    Offline

    11
    ReputationRep:
    (Original post by Molly_xox)
    I thought it asked for standard deviation too!
    It asked for u and sigma
 
 
 
Reply
Submit reply
Turn on thread page Beta
Updated: June 13, 2015

1,065

students online now

800,000+

Exam discussions

Find your exam discussion here

Poll
Should predicted grades be removed from the uni application process

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.