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

    7
    ReputationRep:
    (Original post by money-for-all)
    do we have to show working out for s1 ? or can you just show the answer ? becuase we have graphical calculators
    Where the answer can just be quoted from the graphics calculator, then they will accept it, for example calculating r when given values for x and y.

    If they give summary data for calculations, such as Sxy, Syy, Sxx for calculating r, then you have to show working.
    Offline

    14
    ReputationRep:
    Okay guys, doing some revision and I can't seem to get this question right, it says:

    Given that X~N(20,25) find a when P(X<a)=0.97
    So I used z =(X-u)/s where u is the mean and s is the standard deviation (sorry, I don't know how to get the symbols)
    0.97 = 97% so 97%= P(Z<1.881) so then
    1.881= (a-20)/ 25
    so a = (1.881*25)+20

    Can anyone tell me where I'm going wrong because I don't get the same answer as what's in the back of the book? Thanks in advance!
    Offline

    14
    ReputationRep:
    I don't know how to delete my post but I've just seen where I went wrong!
    Offline

    2
    ReputationRep:
    Sooo true Ive just been doing this and have left it
    Offline

    2
    ReputationRep:
    (Original post by Blake Jones)
    Okay guys, doing some revision and I can't seem to get this question right, it says:

    Given that X~N(20,25) find a when P(X<a)=0.97
    So I used z =(X-u)/s where u is the mean and s is the standard deviation (sorry, I don't know how to get the symbols)
    0.97 = 97% so 97%= P(Z<1.881) so then
    1.881= (a-20)/ 25
    so a = (1.881*25)+20

    Can anyone tell me where I'm going wrong because I don't get the same answer as what's in the back of the book? Thanks in advance!
    I think it's because 25 is the variance normally it would be in the form X~N(20,25^2)
    so try 5 as the standard deviation. Variance = standard deviation ^2
    Offline

    1
    ReputationRep:
    Hello everyone, hope everyone's revision is going well!

    One question... has anyone compiled a list of all the answers that regularly come up? For example, "Give a numerical justification as to why [a variable] is unlikely to be normally distributed" which is always calculating 2<X<4 standard deviation's from the mean. Some questions just have the same answers every time! If there is a list out there... somewhere... We could revise this list and gain the maximum marks possible

    Please and thank you for any help on this x
    Offline

    3
    ReputationRep:
    (Original post by Shaks.smxth)
    Hello everyone, hope everyone's revision is going well!

    One question... has anyone compiled a list of all the answers that regularly come up? For example, "Give a numerical justification as to why [a variable] is unlikely to be normally distributed" which is always calculating 2<X<4 standard deviation's from the mean. Some questions just have the same answers every time! If there is a list out there... somewhere... We could revise this list and gain the maximum marks possible

    Please and thank you for any help on this x
    On that note is there any particular structure on how many standard deviation's from the mean you should use? I've seen it vary between 2 and 3 but I don't know why. Using any proves it if the value comes out negative for both surely.
    Offline

    10
    ReputationRep:
    Hey guys, I was doing some S1 earlier on and I'm just a bit confused at when do you actually use Sx? I get it that it's the unbiased estimator of the variance but when do you actually use it?
    • Thread Starter
    Offline

    7
    ReputationRep:
    (Original post by michelle88222)
    Hey guys, I was doing some S1 earlier on and I'm just a bit confused at when do you actually use Sx? I get it that it's the unbiased estimator of the variance but when do you actually use it?
     s^2 is the variance of a sample, so it is used in sampling, such as for confidence intervals. σ² is the variance of the whole population, so not a sample. The other difference is calculating them, as when calculating the sample standard deviation, you use n-1 whereas calculating the population standard deviation you use n, where n is the size of the sample or population respectively.
    • Thread Starter
    Offline

    7
    ReputationRep:
    (Original post by -jordan-)
    On that note is there any particular structure on how many standard deviation's from the mean you should use? I've seen it vary between 2 and 3 but I don't know why. Using any proves it if the value comes out negative for both surely.
    Usually they allow U - ns, provided that U - ns < 0, which corresponds to either 2 or 3 standard deviations from the mean. I just go for the one which makes it the least negative, as that one is accepted mostly I found.
    Offline

    10
    ReputationRep:
    (Original post by Dapperblook22)
     s^2 is the variance of a sample, so it is used in sampling, such as for confidence intervals. σ² is the variance of the whole population, so not a sample. The other difference is calculating them, as when calculating the sample standard deviation, you use n-1 whereas calculating the population standard deviation you use n, where n is the size of the sample or population respectively.
    Ahh I see. Thanks!
    Offline

    0
    ReputationRep:
    Guys can I ask a silly question.

    I know it says give answers 3sf, but does this apply to both A and B when giving the least squares regression line? and if not, is it better to give both to the same number of decimal places and how many do you recommend?

    An example: A= 1.725550661 and B=0.08480176211 according to the calculator

    So, is it wiser to write my least squares regression line formula as y=1.73+0.08x or y=1.73+0.0848x.
    Is another way better?

    Also, given that the degree of accuracy to which these values are given to in the formula affects the outcome of the predicted y values, is it OK to quote a lower degree of accuracy in the formula than the 700 decimal places the calculator gives, and then just give the calculator values for the predicted y values rather than calculating them yourself?

    Example: The predicted y value where x=23 is 3.675991189 or 3.675 to the 3dp the RESIDUALS at other x values where calculated to ACCORDING TO THE CALCULATOR
    But, USING THE EQUATION Y=1.73+0.0848x as above, you get 3.6804 which of course rounds to 3.680...just that little bit higher

    (sure they round to the same thing at 2dp)

    Does anyone know the general AQA policy on these things, especially whether or not to use your equation to find predicted y values?

    Thanks!!!!

    (The numbers were obtained by crunching those in question 4 of the June 2005 S1B paper )
    Offline

    2
    ReputationRep:
    I've always just done it to 3sf unless stated otherwise, always has worked.

    For the estimating y value, the Ms has a range which includes the calculator estimate and the one u will get for an equation (where a and b are both to 3sf). So either works.
    Offline

    2
    ReputationRep:
    I know its been discussed before about a variable not being normally disturbed as the standard deviations from the mean make the value less than zero meaning that the result is impossible but is it just doing the difference between the mean and how many standard deviations away that confirm this claim?
    Offline

    3
    ReputationRep:
    (Original post by Nai18)
    I know its been discussed before about a variable not being normally disturbed as the standard deviations from the mean make the value less than zero meaning that the result is impossible but is it just doing the difference between the mean and how many standard deviations away that confirm this claim?
    For a bell curve (normally distributed) it is a fact that 95% of values lie within two standard deviations either side of the mean which is the centre. 99% lie within 3. Therefore you can use this to estimate the x-values at both ends and find a range of values which the data lies. Therefore for a quantity such as time, which can't be negative, if you do the mean - 2/3 x the standard deviation and it comes out negative then it's likely not to be normal.
    Offline

    2
    ReputationRep:
    (Original post by -jordan-)
    For a bell curve (normally distributed) it is a fact that 95% of values lie within two standard deviations either side of the mean which is the centre. 99% lie within 3. Therefore you can use this to estimate the x-values at both ends and find a range of values which the data lies. Therefore for a quantity such as time, which can't be negative, if you do the mean - 2/3 x the standard deviation and it comes out negative then it's likely not to be normal.
    Makes sense. Thanks
    Just to be sure, within 3 standard deviations, for your example is it
    mean - 2/3 x standard deviations or is it

    mean - 2 x standard deviations?

    Posted from TSR Mobile
    Offline

    15
    ReputationRep:
    I've only gone through 3/6 chapters and gotta cram tomorrow 😭

    Binomial distribution is similar to normal so that shouldn't be too horrible.

    Then correlation and estimation,

    Anyone think this is possible?

    Hate S1, I'm terrible at if

    Posted from TSR Mobile
    Offline

    3
    ReputationRep:
    (Original post by Dapperblook22)
    The consensus in my class is that June 2013 is the most difficult, it may be different for different classes.
    Thanks! Had a go at it and got 71/75, so I'm feeling confident for tomorrow
    • Thread Starter
    Offline

    7
    ReputationRep:
    (Original post by RickyMaz)
    Thanks! Had a go at it and got 71/75, so I'm feeling confident for tomorrow
    Well done, good luck for tomorrow
    Offline

    0
    ReputationRep:
    Anyone got any last minute tips? I feel like I know it but also want to cram and make sure
    This is a retake for me though so it's a bit easier than last year
 
 
 
  • 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
    Brussels sprouts
  • 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.