Uni mathematics (1 Viewer)

[MysticalElement]

Why is it that practically every university course i look at in UAC says assumed knowledge Mathematics, i mean why do most schools even offer general maths, its basically i waste of time me even doing it (although i am good at it) it seems pointless if it receives no credit......

 

[Iruka]

Assumed knowledge is not the same as a pre-requisite.

They will let you into a degree at uni without the assumed knowledge, but the assumed knowledge will not be taught in that course. Which means that you have to fill in the gaps yourself somehow.

 

[Forbidden.]

Assumed knowledge is not the same as a pre-requisite.

They will let you into a degree at uni without the assumed knowledge, but the assumed knowledge will not be taught in that course. Which means that you have to fill in the gaps yourself somehow.

... and you will struggle if you don't.

 

[darkwolfzx]

In fact one of the guys at a uni open day told me that if I was feeling stupid I could apply for some course that had assumed knowledge of 3 unit when I wasn't even doing it.

Tool.

 

[Affinity]

general maths teaches crappy stats

 

[Affinity]

paraphrasing a statistician: those who learnt "statistics" in HS has to unlearn some of it.

I can probably find a published version.

 

[Iruka]

I taught General maths stats this year and at least half of it is wrong.

The calculus learned in 2 unit would be far more helpful for learning stats at uni.

BTW, I wouldn't get into an argument with Affinity about statistics - he is an actuary.

 

[Iruka]

Yes, I can give you a good example. Go and learn the correct way to calculate an IQR and compare with how you were taught in general maths.

You will find that what you were taught is wrong.

I did not take general maths in school.

Read my post. I said that I have taught it. At school, as a teacher. So I think I am more aware of the syllabus content of General Maths than you.

 

[jas0nt]

IQR = interquartile range

in uni generally first years take quantitative methods which is extremely easy stats. no calculus involved, and 1/3 of the course is hypothesis testing. dead easy stuff.

usually that's where MOST people finish in terms of doing statistics units.

after quantitative methods, i then did regression modelling, which is a mandatory subject for actuary/finance students. no amount of high-school maths will prepare you for this because it doesn't NEED high-school maths, it's an extremely abstract statistics unit which had a midsemester failure rate of 90%.

then comes weird statistics courses like stochastic modelling, where you better prey your calculus is stellar, orelse failure is ensured.

also 90% of the reason why uni's put maths as prereqs is because calculus tends to be involved. essentially every science field uses it, not to mention actuary/economics/fianance. the only subjects that DON'T use calculus i would say would be arts, law and music.

 

[Cookie182]

IQR = interquartile range

Is it still Q3-Q1??? Lol just finished my quant methods midterm and are agreeing with what your saying, its very easy and i can sense that a much harder version lurks in the background.

For eg- I have no idea how, but you know in first year stats with say the Normal Distribution the area under the curve is ASSUMED to be 1. Can anyone here prove this using intergration? Please post it, it would make me happy lol

I did 3U in high school but yea kinda totally forgot the shit lol, i can see 'foreverpink's' point to an extent. I think what she is saying is that if the only stat course ur doing is first year stuff (which is to the level of general maths- commerce only assumes general at my uni) then general gives u a bit of background knowledge. Having not done general, i had no idea about distributions etc but being ok at maths (from 3U) i picked it up easily. Hence, if i was to go on to do later stat courses i would have a competitive advantage over someone who did general (due to my calculus exposure).

 

[Cookie182]

^Bump

I really just posted here cause i wanna see someone prove the area under ND

dunno how difficult this is, havent tried. But it become the main thought of interest whilst i was daydreaming through stat lectures.

 

[Templar]

NEWSFLASH: I didn't have to "unlearn" what I learned back in general - basic probability, the empirical rule, normal distribution have all been pretty much the same as the stats I'm learning now. Besides who is the statistician who said that? Did they do HSC general? That may be the case for the HS that they went to but it may not be the HS's in NSW. And even though that may be the case (which I'm not saying that it is) it isnt that hard to "unlearn" something.

Considering that an emeritus professor who came up with the HSC/UAI scaling system also agree that the statistics taught in high school actually works against uni statistics and that it should be left to university to teach statistics, I would say that has much weight.

 

[BackCountrySnow]

it seems pointless if it receives no credit......

you do it because you want to maximise your UAI.

 

[Yip]

^Bump

I really just posted here cause i wanna see someone prove the area under ND

dunno how difficult this is, havent tried. But it become the main thought of interest whilst i was daydreaming through stat lectures.

Through a substitution of t=(x-muu)/sigma any normal distribution can be written in the form (1/root2pi)e^-(0.5t^2). So we just have to prove that the integral of e^-(0.5x^2) from -infinity to infinity is root2pi.

Letting I=∫[e^-(0.5x^2)]dx=∫[e^-(0.5y^2)]dy (where these 2 integrals are integrated from -infinity to infinity)
I^2=∫∫[e^-0.5(x^2+y^2)]dxdy [integtated over [-infinity,infinity]x[-infinity,infinity]]
Transforming to polar coordinates (i.e. x=rcosa,y=sina),
I^2=∫∫[e^-0.5(r^2)]rdrda=2pi [integrated over [0,2pi]x[0,infinity]] (the extra r comes from finding the jacobian determinant of the transformation)
I=root2pi

 

[Azarnakumar]

Through a substitution of t=(x-muu)/sigma any normal distribution can be written in the form (1/root2pi)e^-(0.5x^2). So we just have to prove that the integral of e^-(0.5x^2) from -infinity to infinity is root2pi.

Letting I=∫[e^-(0.5x^2)]dx=∫[e^-(0.5y^2)]dy (where these 2 integrals are integrated from -infinity to infinity)
I^2=∫∫[e^-0.5(x^2+y^2)]dxdy [integtated over [-infinity,infinity]x[-infinity,infinity]]
Transforming to polar coordinates (i.e. x=rcosa,y=sina),
I^2=∫∫[e^-0.5(r^2)]rdrda=2pi [integrated over [0,2pi]x[0,infinity] (the extra r comes from finding the jacobian determinant of the transformation, if u are not acquainted with this method)
I=root2pi

woah

 

[MysticalElement]

so will i struggle with uni maths even if i took the fundamental maths courses........btw i am quite good at gen maths i aint some dumbshit who fails general

 

[Iruka]

All I can say is that you'd better be prepared to do alot of extra work. I am sure it is not impossible, though. It might depend on what level of uni maths you are planning to take.

 

[MysticalElement]

yeah like i said i'd be planning to take the lowest!! level of maths.....

 

[Cookie182]

Through a substitution of t=(x-muu)/sigma any normal distribution can be written in the form (1/root2pi)e^-(0.5t^2). So we just have to prove that the integral of e^-(0.5x^2) from -infinity to infinity is root2pi.

Letting I=∫[e^-(0.5x^2)]dx=∫[e^-(0.5y^2)]dy (where these 2 integrals are integrated from -infinity to infinity)
I^2=∫∫[e^-0.5(x^2+y^2)]dxdy [integtated over [-infinity,infinity]x[-infinity,infinity]]
Transforming to polar coordinates (i.e. x=rcosa,y=sina),
I^2=∫∫[e^-0.5(r^2)]rdrda=2pi [integrated over [0,2pi]x[0,infinity]] (the extra r comes from finding the jacobian determinant of the transformation)
I=root2pi

wow u r a maths hellman!

Not that i can fully grasp all that (due to lack of knowledge regarding polar coordiantes etc/uni integration) I still thank you for taking the time of proving it.

So im guessing all that boils down to equal 1?

 

[Forbidden.]

CAN'T WAIT TO LEARN PROBABILITY AND STATISTICS IN MATHS 1B IN NEXT WEEK OR TWO.
IT LOOKS BETTER THAN GENERAL MATHS.

Through a substitution of t=(x-muu)/sigma any normal distribution can be written in the form (1/root2pi)e^-(0.5t^2). So we just have to prove that the integral of e^-(0.5x^2) from -infinity to infinity is root2pi.

Letting I=∫[e^-(0.5x^2)]dx=∫[e^-(0.5y^2)]dy (where these 2 integrals are integrated from -infinity to infinity)
I^2=∫∫[e^-0.5(x^2+y^2)]dxdy [integtated over [-infinity,infinity]x[-infinity,infinity]]
Transforming to polar coordinates (i.e. x=rcosa,y=sina),
I^2=∫∫[e^-0.5(r^2)]rdrda=2pi [integrated over [0,2pi]x[0,infinity]] (the extra r comes from finding the jacobian determinant of the transformation)
I=root2pi

Hold up, I'll get it all especially the double integrals next year.
But fuck I forgot whether that first integral converges or diverges.

wow u r a maths hellman!

Not that i can fully grasp all that (due to lack of knowledge regarding polar coordiantes etc/uni integration) I still thank you for taking the time of proving it.

So im guessing all that boils down to equal 1?

ROOT 2 PI