lecture 5

Question 1

what is the procedure for inferential statistics

Answer 1

1. hypthesis
2. significance level: decide on it –> use 0.05
3. sample: take a sample from the population
4. p value: done by a computer –> this is the use of a test like t and chi squared
5. decision has to be made

Question 2

what is a type 1 error

Answer 2

when you reject a true null hypothesis, also called false positive, the probability of type I error is just 𝛼, the significance level or size of test chosen

Question 3

what is a type 2 error?

Answer 3

when you accept a false null hypothesis, also called false negative, the probability of making this error is denoted by 𝛽-depends on the sample size and the population variance

Question 4

what is the formula to find ‘z’

Answer 4

B hat/ SE(Bhat) = p value

Question 5

what is the ‘power’ f a test?

Answer 5

power = 1 - Pr(type 2 error)

stricter the test then the lower the power is

cant attach a specific number to a test but they can be ranked

e.g.
if the tail had an area of 0.088 (5% part), then power = 1-0.088=0.912

Question 6

what is the formula for the z transformation to derive a confidence interval for a slope co-efficient?

Answer 6

Bhat-B/(varience/ E(X-Xbar)^2

Question 7

if you can derive the SE of the intercept, what can you base the tests aroung

Answer 7

Bhat-Bbar/SE(Bhat)

you can do this with the intercept as well (a)