MATH1905 exam (1 Viewer)

[acmilan]

hey all, i dont know if this is a dumb question, but how are we supposed to answer those questions that ask you to do a t-test or a chi-square test considering we arent given the tables? For these past exams i've just been guessing whether the P-value is > or < 0.05 solely by looking at the size of the test statistic :/

 

[gordo]

u sure u arn't given tables?
in 1005 they give us tables

 

[acmilan]

This is all we are given according to the website:

http://www.maths.usyd.edu.au/u/UG/JM/MATH1905/r/formulaNEW.pdf

Just the normal table, you can use that to approximate the binomial distribution, doesnt help with the t or chi-squared one afaik

 

[nit]

You get given tables, similar to the ones at the back of primer. The tutor guy handed sample tables out during the last tute - theyre identical to the primer's tables...we get a normal table i think, as well as percentage point stuff for the t-statistic iirc.

 

[acmilan]

Ah that'd explain it. Thanks

edit: i guess i shouldve done last weeks tutorial, the answer of how to do it is on there

 

[xiao1985]

can't exactly rmber, but i think they actually recycles tables.. so don't be surprized if u see ur table says "this is used by xiao 2004"

 

[fush]

do the tables given out for the 1005 exam include the little graph at the top with the little formula (eg. P(Z < z) = Ф(z) ) cause they are really helpful

 

[tennille]

do the tables given out for the 1005 exam include the little graph at the top with the little formula (eg. P(Z < z) = Ф(z) ) cause they are really helpful

Nope. I remember the lecturer mentioning that.

 

[fush]

damn it ...................

so not even the formula?

 

[tennille]

I don't know about the formula, but I know the graph isn't in it.

Do you know if there are past papers for 1005? I can only find second year stats.

 

[Templar]

Did you check maths or mathematical stats? First year stats is under maths.

 

[tennille]

Yep, found it. It's under mathematics.

 

[LoneShadow]

I was actually hoping there would be no t-test and chi-squared test in the exam as they don't provide us with the tables.................

 

[antarctic]

I remember the lecturer saying something about giving R-outputs.
E.g., if we need the t tables with 9 dof, we'd get the R-output for say 0.01, 0.05, 0.1, 0.25 values for a 9dof t table. So its kinda like getting a portion of the tables in the paper.

Hope that made some sense..

Antarctic

 

[acmilan]

I remember the lecturer saying something about giving R-outputs.
E.g., if we need the t tables with 9 dof, we'd get the R-output for say 0.01, 0.05, 0.1, 0.25 values for a 9dof t table. So its kinda like getting a portion of the tables in the paper.

Hope that made some sense..

Antarctic

Yeah, an example is q1 of the week13 tute

 

[SeDaTeD]

So would it be possible to just fudge your way through but just pull the answer out of the R data?

 

[Templar]

Yeah, the whole stats course is a fudge anyway. Actually, the whole concept of stats is a fudge.

 

[Plebeian]

So would it be possible to just fudge your way through but just pull the answer out of the R data?

The R data is only giving us percentile sort of things:

> q=c(.75,.9,.95,.99)
> round(qt(q,13),3)
[1] 0.694 1.350 1.771 2.650
> round(qt(q,17),3)
[1] 0.689 1.333 1.740 2.567

so you can see where your observed value would go in the distribution. You still have to figure out the observed value yourself ... perhaps the most you could tell is what index the t-distribution has.

 

[SeDaTeD]

Totally bombed this exam .

 

[nit]

same....