ECMT Quiz 3 (1 Viewer)

[stazi]

its ok. all this is to be taken in jest. or jizzed. either one.

 

[04er]

Will someone please tell me how to find P-values???

 

[04er]

i have no idea how to do question 2... and my quiz is in 9 hrs and 20 min... i am so friggen stuffed *sulks*

farout where did we learn how to do this shit? e.g. the anova crap!? ughh

 

[stazi]

sure.
lets say you do the hypothesis test (i'll use fake values).
Ho:u=100
H1:u<100
using z test = 3.55
Now, from what i understand you simply look up the P(Z<3.55) and that is the p-value
I could be wrong though

 

[04er]

sure.
lets say you do the hypothesis test (i'll use fake values).
Ho:u=100
H1:u<100
using z test = 3.55
Now, from what i understand you simply look up the P(Z<3.55) and that is the p-value

thank you stas! how about if we're using the t-values for hypothesis testing? because we're not given population standard deviation... can we just assume that the sample standard deviation is just equivalent to that of the population? and so find z that way? or am i missing something

 

[stazi]

i have no idea how to do question 2... and my quiz is in 9 hrs and 20 min... i am so friggen stuffed *sulks*

farout where did we learn how to do this shit? e.g. the anova crap!? ughh

ok. add me to msn: stas_m_@hotmail.com
if i knew yesterday, i couldve met up with you today and taught you all the anova crap. its really necessary and difficult to just explain.
I'll help you a lot, if you in exchange tutor me for probability some time down the track

 

[04er]

ok. add me to msn: stas_m_@hotmail.com
if i knew yesterday, i couldve met up with you today and taught you all the anova crap. its really necessary and difficult to just explain.
I'll help you a lot, if you in exchange tutor me for probability some time down the track

haha well i can definitely help you out with what i know... are you on msn atm?

 

[stazi]

thank you stas! how about if we're using the t-values for hypothesis testing? because we're not given population standard deviation... can we just assume that the sample standard deviation is just equivalent to that of the population? and so find z that way? or am i missing something

Yes, you are correct. we need to use the t value. Unfortunatelly the tutor tried to explain it and he confused us sooo damn much it was crazy.
However, the text just says that we use excel to find the p-value of a t test.
I have no idea how to do it.
I also feel sorry for whom get sthe type 2 error Q.

 

[sarevok]

When n (greater than or equal to 100), the Z table approximates the T table. Thus, you can find the P-value by looking up P(tobs<Z) or P(tobs>Z) depending on where the rejection region is. If n<100, it becomes tricky, you'd have to make some crude approximation by looking at the t tables.

 

[stazi]

When n (greater than or equal to 100), the Z table approximates the T table. Thus, you can find the P-value by looking up P(tobs<Z) or P(tobs>Z) depending on where the rejection region is. If n<100, it becomes tricky, you'd have to make some crude approximation by looking at the t tables.

my tutor says that we shouldnt really use the Z table, even to approximate, as the t table gives you the infinite value = z figures. or something like that.

 

[myg0t]

for parts of question 1) as 1 time didnt know as well

how do you work out the probability of a type 2 error? for example in question 1d part iii)

cheers

Av my good man.

Probability type 2 error is that you will get 1/10 today.

Probability type 1 error is that i will get 2/10 today.

 

[kow_dude]

is anyone doing their quiz on monday ?

 

[5c]

yeh... me me!!
n i fink i'll be screwd tomrw

 

[polska*]

i am and not looking forward to it. i havew no clue as to what is going on

 

[stazi]

i won that quiz big time.
you guys are going to get the hardest questions.
the residual analysis and type 2 probability. Oh and p-value of a t test.
No pressure.

 

[kow_dude]

i won that quiz big time.
you guys are going to get the hardest questions.
the residual analysis and type 2 probability. Oh and p-value of a t test.
No pressure.

Pfffft........

 

[5c]

not sure if dis was answerd...
but how do we find: (in c)
I=?
J=?
[the table]
in the first question of the regression q's...

please explain~~

 

[stazi]

what does I and J represent? dont have the quiz with me.

 

[5c]

the table:

for the intercept row....
coefficient
I

standard error
13.835

t stat
18.834

for the X row,
coefficient
-22.969

standard error
J

t stat
-6.457

 

[stazi]

so I is the intercept coefficient?
If so intercept coefficient is just Standard error * t stat
K is the p-value I presume. We can estimate it at 0, similarly to the above p-value.
This can partially be calculated from the hypothesis test we did. I think there's another reason which I can't remember. But generally if theres a large slope, then p = 0.0000....