The Student Room Group

s3 confidence levels test help needed

https://i.imgur.com/smd06TG.jpg
for part a is it correct to say that
h0: mu =0.5
H1=mu not = 0.5
and since 0.5<0.5127 at 10% significance we accept h0 ???
sorry I don't have ms so cannot check.
Reply 1
I would have said that 0.5 lies inside the confidence interval therefore accept H0.
What you say should be fine.
Original post by vc94
I would have said that 0.5 lies inside the confidence interval therefore accept H0.
What you say should be fine.

thank you
https://i.imgur.com/8qV7lJe.jpg
is the question stating that U{0,10}
is a uniform distribution????
hence how would I find the expected frequency
@vc94
Original post by assassinbunny123
thank you
https://i.imgur.com/8qV7lJe.jpg
is the question stating that U{0,10}
is a uniform distribution????

@ghostwalker
Original post by assassinbunny123


Yes, it's a uniform distribution.
Original post by ghostwalker
Yes, it's a uniform distribution.

how would I find the expectations then?
Reply 7
Original post by assassinbunny123
how would I find the expectations then?


Assuming H0 is true, so X~U[0, 10], then, for example, P(0<X<4) = 0.4, so expected freq for this class is 0.4xtotal freq = 0.4x100= 40

Do the same for the 3 other classes.
Original post by vc94
Assuming H0 is true, so X~U[0, 10], then, for example, P(0<X<4) = 0.4, so expected freq for this class is 0.4xtotal freq = 0.4x100= 40

Do the same for the 3 other classes.

did you find 0.4 from using the probability density function 1/b-a and then multiplied it by the radius?
Reply 9
Original post by assassinbunny123
did you find 0.4 from using the probability density function 1/b-a and then multiplied it by the radius?


P(0<X<4)= 4/(10-0) = 0.4 from the pdf
Original post by vc94
P(0<X<4)= 4/(10-0) = 0.4 from the pdf

thank you very much i understand this now.
one more question how would i find the expectation freq for a normal distribution?
Reply 11
Class boundaries would be:
X<6.5
6.5<X<7.5
etc..
10.5<X<11.5
X>11.5

Multiply probabilities by 175 to get Expected freqs.
Original post by vc94
Class boundaries would be:
X<6.5
6.5<X<7.5
etc..
10.5<X<11.5
X>11.5

Multiply probabilities by 175 to get Expected freqs.

would we use the midpoints of the class boundaries ?
I don't think its necessary to use midpoints. Just work out the probability that a<X<b (e.g. 6.5<X<7.5) for each respective class in the normal way (i.e. using the z statistic) then multiply that probability by the total number of trials to work out the expected value.What past paper is that from?

Quick Reply