The Student Room Group

Estimators - Statistics

For estimators, a good measure of their efficiency is the standard error (S.E.) of the estimate. does this apply to any estimate or just estimates of the mean, and is it always the variance that is used in determining the S.E.?

if you have an estimate of the mean and for this population of size n, you know:

the population variance
have calculated a (unbiased~n-1 denom)sample variance

How do you determine the standard error of the mean from each of the (distinct)situations above

ive just been confused by my book about dividing by sqrtn etc.

Aswell as the formula, could anyone give some sort of an explanation for it in each case,as like i have said, ive become confused about this notion

cheers
Hi there,

While you're waiting for an answer, did you know we have 300,000 study resources that could answer your question in TSR's Learn together section?

We have everything from Teacher Marked Essays to Mindmaps and Quizzes to help you with your work. Take a look around.

If you're stuck on how to get started, try creating some resources. It's free to do and can help breakdown tough topics into manageable chunks. Get creating now.

Thanks!

Not sure what all of this is about? Head here to find out more.
Reply 2
Bump
Original post by newblood
For estimators, a good measure of their efficiency is the standard error (S.E.) of the estimate. does this apply to any estimate or just estimates of the mean

Don't know. But probably. Question is too general for me.


, and is it always the variance that is used in determining the S.E.?


You may find this useful.
Stop reading when you get to "correction for finite population" label - it's not relevant to A-level, and will only confuse.


if you have an estimate of the mean and for this population of size n, you know:

the population variance
have calculated a (unbiased~n-1 denom)sample variance


Variance uses "n", whether you're dealing with random variables, or with samples.

When you want an unbiased estimate of the variance of a population, given a sample then you use "n-1"

Hope that helps.
Reply 4
Original post by ghostwalker
Don't know. But probably. Question is too general for me.



You may find this useful.
Stop reading when you get to "correction for finite population" label - it's not relevant to A-level, and will only confuse.



Variance uses "n", whether you're dealing with random variables, or with samples.

When you want an unbiased estimate of the variance of a population, given a sample then you use "n-1"

Hope that helps.


Cheers. On a slightly unrelated note.

If you have a 3- dimensional vector given in component form

How would you find the modulus-argument form (r,theta)

I cant see how to work out theta
Original post by newblood
Cheers. On a slightly unrelated note.

If you have a 3- dimensional vector given in component form

How would you find the modulus-argument form (r,theta)

I cant see how to work out theta


In three dimensions you need three coordinates. There is more than one "standard" form.

First ask do you need this? What's it for? I don't recall this being covered at A-level.
Reply 6
Original post by ghostwalker
In three dimensions you need three coordinates. There is more than one "standard" form.

First ask do you need this? What's it for? I don't recall this being covered at A-level.


Its not come up in my alevel but its quite a basic thing in alevels to be asked to convert a 2-D vector into modulus-argument form. So i was just curious about how to extend this idea to3d. The 2d idea is pretty simple so i assumed, naturally, that the same might be true for an i,j,k vector
Original post by newblood
Its not come up in my alevel but its quite a basic thing in alevels to be asked to convert a 2-D vector into modulus-argument form. So i was just curious about how to extend this idea to3d. The 2d idea is pretty simple so i assumed, naturally, that the same might be true for an i,j,k vector


With 2D, you only need to specify a distance from the origin and an angle relative to the positive x-axis (that being the usual method), and your point is defined.

In 3D you need three coordinates. You can specify a distance from the origin, but you then need two angles (or an angle & a displacment).

Have a look at spherical coordinates and cylindrical coordinates on wiki - those are the two (three actually as there are two flavours of spherical) main ones.

But I wouldn't concern myself with them in any detail until you have the A-level stuff down solid.
(edited 9 years ago)
Reply 8
Original post by ghostwalker
With 2D, you only need to specify a distance from the origin and an angle relative to the positive x-axis (that being the usual method), and your point is defined.

In 3D you need three coordinates. You can specify a distance from the origin, but you then need two angles (or an angle & a displacment).

Have a look at spherical coordinates and cylindrical coordinates on wiki - those are the two (three actually as there are two flavours of spherical) main ones.

But I wouldn't concern myself with them in any detail until you have the A-level stuff down solid.


thanks! ill have to have a look at that later down the line then

Quick Reply

Latest