ECMT Quiz 3 (1 Viewer)

[sarevok]

Ugh, I had no idea we had this quiz so soon after the last one.

Potential questions will be up tomorrow, apparently.

 

[04er]

woopeedoo

 

[absolution*]

fuck

long questions :S

 

[myg0t]

Looks like MDS will have the last laugh afterall....

F!ck hypothesis testing, f!ck regression ... f!ck you Murray!



0/10 here we come...

 

[stazi]

what the fuck!
question 1 is awesome.
The rest looks like murrays toilet bowl in the morning.
and i dont want to 'work the paper' as it suggests.

 

[stazi]

here's my workings to q1.a.
can someone double check im doing the right thing, if so, then ill post more

 

[kow_dude]

omg... starting already ? Incredible.....

 

[s.m.i.t.h]

when do u know to use one tail or two tail test?

 

[Komit]

I think you're supposed to use the z-test since the standard deviation is known........

 

[stazi]

when do u know to use one tail or two tail test?

1) kowdude: I have a marketing presentation thursday. Want to get this shite over with.
2) Smith: when the alternate hypothesis is * x

3) Komit: the standard deviation (σ) is unknown, only the sample stdev is known (s)

 

[Komit]

Ahhhh. I see.

 

[stazi]

can someone confirm that my answer is right. if it is ill try finish q1 by tonight.

 

[Lainee]

Based on lecture notes and notes from my workshop (as opposed to the textbook) I would say:
Ho: u=50000
H1: u>50000

 

[stazi]

but its saying:
The average km should be greater than 50,000
so were testing if theyre less than.

I'm not sure why they would test if its greater than. Logically, also theres a risk if the company says tires last on avg 50k km when they last less as opposed to the other way round.

 

[011]

Testing less than and finding a rejection and testing greater than and finding a support of the null are equivalent. You can do what you want with it.

 

[stazi]

question 1b

 

[stazi]

hmm it doesnt say how many q's we'll get.
doing 1 hypothesis test (if set out properly like im doing it) takes 5-10 mins. There's a lot of writing involved.
1c.

 

[stazi]

ok, here's where i get confused.
am i doing the p-test right?
How the hell do you calculate the progaybility of a type 2 error?

 

[stazi]

p-values: bad again
probability of type 2 error: no idea again.

 

[stazi]

pvalues: again poop??
type 2 prob: again poop.