Competition - post a photo you've taken of nature >>

AQA Statistics 2 (B) Exam Thursday 21st June

Scroll to see replies

Hey, I don't know if this is helpful or not, but I always get confused as to whether to use n, or n-1, and I found this guide that the exam board publish as to which to use; http://store.aqa.org.uk/qual/pdf/AQA-6360-6380-W-TRB-S1-GUIDANCE.PDF
Might be worth a look :smile:

I just looked at it and it didn't make that much sence, the way I understand it is for sample variance divide by n-1 and for the variance divide by n...
Is this correct?

Reply 41

5elsste

Original post by Miyata

Can someone please explain 6c) june 2011? http://store.aqa.org.uk/qual/gce/pdf/AQA-MS2B-W-QP-JUN11.PDF can't do it, don't understand what they even want. Everything was going so well, but 6c... -6 marks, and last question, -7 marks, ran out of time so and messed up...

For 6c You know the top half equation of the cumulative distribution is 0.5, therefor integrate the other equation between q and 1 and make it equal to 0.25... Simplify and you will end up with a quadratic, solve and one of your results will lie outside the inequality so it has to be the other.

Reply 42

Miyata

Original post by 5elsste

Sorry, my mistake I meant 4c) both parts i) and ii). I managed to do 6c, but not d), I am thinking about d, almost understood it, but I am concerned about 4c I don't get it at all. Sorry again, hopefully you could explain 4c to me.

Reply 43

PhysicsGirl

Original post by 5elsste

I just looked at it and it didn't make that much sence, the way I understand it is for sample variance divide by n-1 and for the variance divide by n...
Is this correct?

Essentially, yes :smile:

But it was saying that if asked for sample variance, you're allowed to be credited for using either, whereas if you're asked for an unbiased estimator of sample variance, you're have to use n-1, and that if you had to use sample variance indirectly (for example as part of a hypothesis test), you need to use n-1, but would receive full marks for using n iff n>30. :smile:

Reply 44

ailish3park

Hey, could someone please explain how to do stuff for continuous distribution, E(X), Var(X), median, quartiles etc and rectangular distribution. I'm not understanding anything the book is saying. Thank you =)

Reply 45

OnimushaGTA

Anyone can help me here?

For the hypothesis testing, I can do the maths..etc but why do you accept the conditions (let's say H0) and reject the other?!

The only thing which I can't do atm!

Muchos gracias

Reply 46

OnimushaGTA

Original post by ailish3park

For the continuous distribution you have to integrate the function given!

Let's say the question says

f(x) = 1/3 2<x<4

To find E(X), you have to integrate 1/3 multiplied by x between the limits of 2 and 4; does that makes sense?

For Var(X), remember it is E(x^2) - E(x)^2 and work out your answer from there!

(for E(x^2) -basically integrate 1/3 multiplied by x^2 [as it is E(x^2)])

Does this makes sense?!

Reply 47

Lollyage

Original post by OnimushaGTA

You reject the null if the value you get exceeds the critical value :smile:

Reply 48

Nerdatious

Original post by Lollyage

Why didn't I take Decision 1 this year?

have a group snug to comfort you...I need one too really!

Reply 49

OnimushaGTA

Original post by ailish3park

Median, you should know from S1 that this is the middle number?
Quartiles - lower quartile 25% so 0.25; upper quartile 75% so 0.75 and so on!

Rectangular distribution is best illustrated by this (think of continuous random variables):

f(x) = 1/b-a a<x<b

To work out the mean it is 1/2(a+b) and the variance is 1/12(b-a)^2

Apart from these differences, it works similarly as a continuous distribution model; thus integration..etc! (could show you a proof but v. long + not required)

(edited 11 years ago)

Reply 50

ailish3park

Original post by OnimushaGTA

Yeah kind of, thank you!

Reply 51

OnimushaGTA

Original post by Lollyage

You reject the null if the value you get exceeds the critical value :smile:

LALITA!

Haha! Sometimes that's not always the case though? I think :s-smilie:

Some questions I did, where the value does not exceed the critical value and it was rejected!

Maybe I misread it :P

Reply 52

Lollyage

Original post by Nerdatious

have a group snug to comfort you...I need one too really!

Aww

Massive snug is due after exams :yep:

Original post by OnimushaGTA

LALITA!

Haha! Sometimes that's not always the case though? I think :s-smilie:

Some questions I did, where the value does not exceed the critical value and it was rejected!

Maybe I misread it :P

NILESH!
It gets kinda confusing sometimes with negative numbers, so you have to be extra careful with those, but as far as I'm aware this rule always applies. Make sure you write the H0 and H1 correctly in the first place too, that's where I often screw up :p:

Reply 53

OnimushaGTA

Original post by Lollyage

Aww

Massive snug is due after exams :yep:

Okie dokes, need to do some more questions :P

(currently having roti and chicken curry :wink:

)

Reply 54

OnimushaGTA

Original post by ailish3park

Yeah kind of, thank you!

Any more questions, quote me, I'm more than happy to help!

Reply 55

Lollyage

Original post by OnimushaGTA

Okie dokes, need to do some more questions :P

(currently having roti and chicken curry :wink:

)

Haha, funnily enough I'm having chicken curry tonight too :colone:

Comfort food before the dreaded S2 :yep:

Reply 56

coolstorybrother

So I know that P(a⎪b) = P (a ∩ b) / P (b)
However in this question 4 cii (below) to my understanding - i may be wrong :L it does not this, to my understanding they are discrete random variables which they have done P (a) / P (b) as shown in the mark scheme
http://store.aqa.org.uk/qual/gce/pdf...W-QP-JAN11.PDF

Mark scheme:

http://store.aqa.org.uk/qual/gce/pdf...W-MS-JAN11.PDF

However for continuous random variables question 6d they have done :
P (a ∩ b) / P (b)

Why is this the case? Is P(a⎪b) for Discrete random variables P (a) / P (b) and for continuous P (a ∩ b) / P (b)? Please explain if i am wrong thanks