# Tips for S2 exam!!!!!!Watch

This discussion is closed.
#21
ur right emma but don't the distribution questions always have to be to 4dps?
0
14 years ago
#22
Before you enter the exam, you should:

Have a break, have a kit kat.
0
14 years ago
#23
ur right emma but don't the distribution questions always have to be to 4dps?
For your working out you would most definitely use more than 3 significant figures, however for your final answer you only have to give it to 3 sf, unless otherwise stated.

CLT = Central Limit Theorem, if the original distribution follows a normal then you can use the CLT regardless the size of n.
0
14 years ago
#24
i always gine answers to 4 dp because all the probaliities in tables are given to 4dp
0
14 years ago
#25
When sketching a pdf/cdf, make the lines along the x-axis nice and thick, or marks will be lost. AND DON'T BE STUPID LIKE ME, LABEL THE AXES.

like if it says f(x) = 0 for x < 1 then there has to be a thick line going from 1 to negative x along the x-axis (my teacher, who used to mark a-level papers, told me this).

If you're doing a question where X~B(n,p) or X~N or or X~Blah de blah
always write what x actually is.

Learn the definitions of things like sampling distribution, statistic, population, samlping unit, etc, learn the conditions for binomial and poisson - there's always some kind of wordy question.

Good luck everyone!
0
14 years ago
#26
Continunuity corrections are a biggy too.
0
14 years ago
#27
Another thing I just remembered, which I wouldn't have known if I hadn't seen it on a past paper markscheme:

The value of p that will give the most accurate estimate when approximating the binomial distribution by a normal distribution is p = 0.5, as this will give a symmetrical binomial distribution.

Usually, I think the "Approximation Hierarchy" goes something like:

Binomial can be approximated by Poisson for small values of p and large values of n, and this is generally the method to use. If this is not practical for the question (i.e., required values not in the table), then Normal can be used (providing np>5 and np(1-p)>5.

Poisson can be approximated by Normal for lambda > 10.

Disclaimer: I'm quite sure this is right, but if it's not, may I never eat ice-cream again.

Dagnabbit, another thing - when approximating using Normal, don't forget to change the values accordingly, like

X~B(n, p)
P(X<12)
becomes
P(Z<11.5) where Z~N(np, np(1-p))
0
14 years ago
#28
and don't forget to minus the square of the mean, I always do that on variance of a pdf...
0
X
new posts

Latest
My Feed

### Oops, nobody has postedin the last few hours.

Why not re-start the conversation?

see more

### See more of what you like onThe Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

### University open days

• University of East Anglia
All Departments Open 13:00-17:00. Find out more about our diverse range of subject areas and career progression in the Arts & Humanities, Social Sciences, Medicine & Health Sciences, and the Sciences. Postgraduate
Wed, 30 Jan '19
• Aston University
Wed, 30 Jan '19
• Solent University
Sat, 2 Feb '19

### Poll

Join the discussion

Remain (753)
80.36%
Leave (184)
19.64%