Edexcel S3 - Wednesday 25th May AM 2016

Scroll to see replies

Yeh just did june 15 S3, cudnt do the 6 marker Q5 so yeh. Other then that the paper is easy lol.lets ee if i got everything else right. Only need 70 so im not too bothered haha.
Took 1hr 10 mins overall. Slow but i aint no statistician.

Reply 941

Zacken

Original post by Ayman!

I got part a and why the mean is a + 2. I'm uncertain as to how we're suddenly using the sample mean to find a's CI in part (b). I thought you're supposed to use the population mean?

It would have been lovely to find the population mean, but we don't know what the population mean is. Sure, the population mean is distributed with a mean. We're given a sample and with that sample we estimate the population mean that then allows us to estimate a.

Reply 942

tazza ma razza

Original post by paradoxequation

Anyone know which papers have a question where you need to test either the normal distribution or continuous uniform distribution as a suitable model? Thanks.

http://www.madasmaths.com/archive/maths_booklets/statistics/goodness_of_fit.pdf

keep scrolling and the last few qns

Reply 943

Rkai01

https://771a1ec81340d97ae9ed29694f73dd633b1c7c70.googledrive.com/host/0B1ZiqBksUHNYV1BTbDkxWXZCVmc/CH3.pdf

Ex b q 6 a for shouldn't it be 250p as we'd expect that many sixes over the course of 25 days?

Reply 944

username1162972

Decent, got 66/75 in June 15, not bad at all for me.
Cudve got higher if i say 'Not significant' instead of 'insufficient eveidence to reject H_0' in the harry part. I had no clue on the standard error question since i have no understanding of statistics whatsoever 😂

Posted from TSR Mobile

Reply 945

connorbarr

Original post by physicsmaths

say both no? result is insignificant therefore insufficient evidence to reject null hypothesis

Reply 946

Krollo

Original post by physicsmaths

Not bad. I got 73/75 on that one this morning, we'll smash this bro. I have a feeling it'll be an easyish standard paper like s3.

Posted from TSR Mobile

Reply 947

Krollo

Just going to do 2015 IAL now as my last mock.

Posted from TSR Mobile

Reply 948

kennz

When do we use the critical values for a chi squared test that are for significance levels of 90%,95% etc

Reply 949

connorbarr

https://869d950bf149eefd5cd2652b400160ceb38ca5c7.googledrive.com/host/0B1ZiqBksUHNYVGVDMmVJYzVnRnc/June%202009%20QP%20-%20S3%20Edexcel.pdf for question 2a and 2b, what is the difference between what they are asking apart from one says confidence interval and they have different levels?

Reply 950

L'Evil Wolf

https://771a1ec81340d97ae9ed29694f73dd633b1c7c70.googledrive.com/host/0B1ZiqBksUHNYV1BTbDkxWXZCVmc/CH4.pdf

On page 44 how come they have used different starting and ending intervals to the one in the table?

Reply 951

username1162972

Original post by Krollo

Not bad. I got 73/75 on that one this morning, we'll smash this bro. I have a feeling it'll be an easyish standard paper like s3.

Posted from TSR Mobile

indeed mate. this module is ridiculously easy, one should not be able to learn majority of the module in 3 days and get a*s in papers it just isn't fair haha. i have only done 2 full papers haha.
also generally, what does standard error represent?
i was really hoping for a hellish m3 since i knew that anything they throw i would be alright with. i could vary from 75-100 on s3 lets see wht happens haha.
even if all goes pear shaped i can get 90s on m45 and still turn out for the a*

Reply 952

Kevin De Bruyne

Original post by connorbarr

I would interpret 2a as just finding the endpoints of the confidence interval without expressing it as an interval.

The second is asking for an actual interval, with a different confidence level as you said.

Is there something in particular that you're looking for?

Reply 953

MarcHughes

Original post by kennz

When do we use the critical values for a chi squared test that are for significance levels of 90%,95% etc

It might help if you look at exercise 4A in the S3 Edexcel book. It's basically only used if they ask for it.

Reply 954

connorbarr

Original post by SeanFM

for part a mark scheme has done exactly what you would do if it asked for confidence interval except instead of standard error, it has just done 0.5 so just SD dont really get it

Reply 955

Kevin De Bruyne

Original post by connorbarr

for part a mark scheme has done exactly what you would do if it asked for confidence interval except instead of standard error, it has just done 0.5 so just SD dont really get it

Ohhh, of course, sorry, there is another key difference - I see what you were asking now.

a) asks for the distribution of the orchids themselves, not the distribution of the mean of them.

So you use that a population is distributed ~N(20.1,0.5^2) for a) and a mean of n samples is distributed ~N(20.1, 0.5^2/n) for b).

Reply 956

tazza ma razza

is a discrete uniform basically a pdf?

Reply 957

Armpits

For goodness of fit tests, say you have a binomial distribution, do you work out the parameter p by default? Or do you just assume it. E.g. in an event with two outcomes, do you assume p is 0.5, or calculate it from observed data? I calculated in one q and lost some marks as it was supposed to be assumed.

Reply 958

Geraer1

Original post by Ayman!

I got part a and why the mean is a + 2. I'm uncertain as to how we're suddenly using the sample mean to find a's CI in part (b). I thought you're supposed to use the population mean?