The Student Room Group

Scroll to see replies

Reply 40
can anyone shed some light on why (in the jan 08 s2 paper)

in part 8d
m^2 – 4m + 4 = 0.5
where did the +4 come from??

(the question is to find the median of:
f(x)=2(x-2) 2<x<3 )
Thanks guys
error502
can anyone shed some light on why (in the jan 08 s2 paper)

in part 8d
m^2 &#8211; 4m + 4 = 0.5
where did the +4 come from??

(the question is to find the median of:
f(x)=2(x-2) 2<x<3 )
Thanks guys


2m2(x2) dx=0.5\displaystyle\int^m_2 2(x-2)\ dx = 0.5

[2x224x]2m=0.5\left[ \frac{2x^2}{2} - 4x \right]_2^m = 0.5

[(m24m)(224(2))]=0.5[(m^2 - 4m) - (2^2 - 4(2))] = 0.5

(m24m)(48)=0.5(m^2 - 4m) - (4 - 8) = 0.5

(m24m)(4)=0.5(m^2 - 4m) - (- 4) = 0.5

m24m+4=0.5m^2 - 4m + 4 = 0.5
Reply 42
Liverpool F.C.™
Yes, you can. How do you know which one is better ? (Without working out the actual Binomial distribution). My conditions satisfy both Poisson and Normal when choosing if it is a good approximation.


If you look in the S2 book (new one) on page 89, there is a paragraph which explains in which case to either go to normal or poisson, it does say in the paragraph that if you are not sure of which approximation to use, then calculate np, and if it is less than or equal to 10 then the poisson approximation should be used, but if it is greater then 10, then normal approximation should be used.
Reply 43
when finding the median do you always have to intergrate with limits of m and the lower limit?
Reply 44
himegan
when finding the median do you always have to intergrate with limits of m and the lower limit?


Well the whole point is to find the + C after you integrate, you can do it Liverpool F.C's way, or just integrate and you know F(x) = 0 when x = lower limit, so jus sub them in and find the +C, then make F(x) = 0.5, and then find your median. But Liverpool F.C's way works fine, as you know the area between the lower limit (2) and the median (m) is 0.5, so if you integrate f(x) using limits m and 2, it will equal 0.5, also this works for upper quartile and lower quartile only you equate it to 0.75 or 0.25 depending on what quartile you are finding.
Reply 45
thank you!
Reply 46
sorry to be a pain...

i understand Liverpool FC's method. But, when i try to do this again, with the 'normal' method of trying to find the integral of f(x), then equate that to 0.5, I CANT FIND C!?!? when i substitute the upperbound i get c=3, and when i substitute the lowerbound i get c=4.

f(x)= 2x-4 2<x<3

what am i doing wrong??
thankyou!
himegan
when finding the median do you always have to intergrate with limits of m and the lower limit?


Not always, when you have the c.d.f, F(x) = 0.5 since the c.d.f. is the integrated function of the p.d.f.

If you're given a p.d.f. you must integrate first.



Think of it this way, essentially the median is when the probability less then or equal to a certain value is equal to 0.5, P(X &#8804; m) = 0.5. Since we are finding an area under a function we must integrate, the lower limit must the lowest value and since we don't know the upper limit to make the area 0.5, we just label it m.
error502
sorry to be a pain...

i understand Liverpool FC's method. But, when i try to do this again, with the 'normal' method of trying to find the integral of f(x), then equate that to 0.5, I CANT FIND C!?!? when i substitute the upperbound i get c=3, and when i substitute the lowerbound i get c=4.

f(x)= 2x-4 2<x<3

what am i doing wrong??
thankyou!


Can you show me exactly what you are doing ? I'm not getting what you're trying to say.
Reply 49
error502
sorry to be a pain...

i understand Liverpool FC's method. But, when i try to do this again, with the 'normal' method of trying to find the integral of f(x), then equate that to 0.5, I CANT FIND C!?!? when i substitute the upperbound i get c=3, and when i substitute the lowerbound i get c=4.

f(x)= 2x-4 2<x<3

what am i doing wrong??
thankyou!


If I were you, I would stick to Liverpool FC's way if you are confused about the way I told you, look at the post above he fully explains it. Otherwise here is the way I usually do it:

First lets find F(x), so integrate f(x) to get x^2 - 4x + C

so F(x) = x^2 - 4x + C

we know F(x) = 0, when x = 2, so (2)^2 - 4(2) + C = 0, C = 4 (this is where you went wrong, you equated it to 0.5 to find +C, sorry if i wasn't clear on it,)

so F(x) = x^2 - 4x + 4, now we equate it to 0.5 to find the median

x^2 - 4x + 4 = 0.5, and then just find x
Reply 50
My bad! forgot to put F(x)=1, when using the upperbound! left it as F(x)=0!!

okay, so now iv got c= 4

shouldn't i be able to equate F(0.5)= median??

so, x^2 -4x + 4 =median where x=0.5

which comes out as median=2.25
the mark scheme says it's 2.71 though
Reply 51
okay! GOT IT! so equate F(x) to 0.5, not x=0.5!!!!!!!!!!!!!!!!!! THANK YOUUU!
error502
okay! GOT IT! so equate F(x) to 0.5, not x=0.5!!!!!!!!!!!!!!!!!! THANK YOUUU!


Be careful not to confuse the p.d.f and c.d.f. in the exam. Very important, and could lead you to get the whole question wrong.
Reply 53
thank you, help much appretiated!
Reply 54
error502
My bad! forgot to put F(x)=1, when using the upperbound! left it as F(x)=0!!

okay, so now iv got c= 4

shouldn't i be able to equate F(0.5)= median??

so, x^2 -4x + 4 =median where x=0.5

which comes out as median=2.25
the mark scheme says it's 2.71 though


Its F(0.5) not x=0.5!!
Reply 55
I've gone through all the edexcel papers. I was just wondering if you think it would be helpful to go through OCR and AQA papers as well... opinions anyone??
Anyone else finding it hard to remember all of the Populations and Sample definitions? Ugh.

That might just be because im revising for 4 Maths exams at the same time, but it's still a bitch.
TwilightKnight
Anyone else finding it hard to remember all of the Populations and Sample definitions? Ugh.

That might just be because im revising for 4 Maths exams at the same time, but it's still a bitch.


Snap on both accounts!

All the words are the same/similar and none of it's sticking :frown:

As for the insane amount of maths exams.... :sigh:
Oh bugger I've forgotten to learn those definitions! MUST remember to tomorrow...
Reply 59
Liverpool F.C.™
You can approximate using a Normal. (Since np>5 and nq>5)

So a binomial distribution X~B[n,p] &#8776; X~N[&#956;,&#963;²]

Where &#956; = np; &#963;² = np(1-p)

So in your case that would be X ~ N[10,25/3]

P(X&#8805;16) Using Continuity correction would be P(X>15.5)

Standardising it: P(Z>15.510253)P(Z>\frac{15.5-10}{\sqrt {\frac{25}{3}}})

So P(Z>1.91) (Answer is to 3 sig. fig.)

1 - P(Z<1.91)

Using tables P(Z<1.91) = 0.9719

&#8756; 1 - 0.9719 = 0.0281


right, but can i find it throught the binomial method?

Latest

Trending

Trending