Let Y~T-student(v=18),Find a)P(Y<=0.15) and P(0.025
I'll be back for this :)
Is it about z score, area?
it follows t-student distribution
just this informations
I'll get you better help because I can't even trust myself with this field !
ok thanks
@jim_thompson5910 Would you give us a hand, thanks!
are you familiar with student T distributions?
not much
alright one sec while I get a table
alright have a look at this table http://www.sjsu.edu/faculty/gerstman/StatPrimer/t-table.pdf
ok
Have you see something like this before?
nope can you explain me the parameters?
The parameter that will define which values we're working with is the degrees of freedom, which in this case is 18
ok
so you only focus on the values that are in the row that start with 18
ok
now up at the top, where it says "cum. prob", "one tail" "two tail", do you see that?
yes
alright, we'll only be focusing on the "one tail" row
ok but why?
I'm just now realizing that there are holes in this table, one sec
unfortunately, the only tables I've found rely on using them backwards, but the problem is that they don't even come close to the values being used here.
are you able to use a calculator instead?
yes,but could you explain how to use that table with other vaues,my cal do that. but i have to explain how i find the values in test
i have hp 50g
Well to calculate P(Y<=0.15) for instance, you have to look for a 0.15 in the table or something close to it...but...you're confined to the 18th row. So you only have 4 or 5 shots to get it right. So you can see this is far from accurate. Example: If you wanted to calculate say P(Y > 1.067), then you would look for 1.067 which is the fourth element in the 18th row and you would look at the value in the "one tail" row right above 1.067 to see the value 0.15 So P(Y > 1.067) = 0.15 So if you're given values that land on or near the values in the table, then that's great. Otherwise, you're in a lot of trouble.
what means t.85?
that's the 85th percentile
ok
ie, P(Y < 1.067) = 0.85 or 85% of the distribution lies below t = 1.067
and that z ?
so you have to explain how you got these t cdf values?
as v --> infinity, the t-distribution approaches a normal distribution
which is N(0,1)
basically, for really large degrees of freedom, you can use the standard normal distribution (instead of t-distributions)
so they're just showing how it's connected here
ok thanks a lot jim
you're welcome
Join our real-time social learning platform and learn together with your friends!