slide4 - Hypothesis testing example An automotive company...

Info iconThis preview shows pages 1–10. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Hypothesis testing: example  An automotive company claims that its new motorcycle  produces 145 hp at the rear wheel. It is known that the  standard deviation for the whole population is equal to 3  hp.  Let’s take several size 10 samples and try to find out  whether the statement can be trusted. Given a value of the sample mean suggest a criterion  according to which the claim (hypothesis) can be  rejected. How easy is it to reject a true claim?
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 Definitions Critical region : range of values of the  test statistic for which the null  hypothesis is rejected Acceptance region : range of values of  the statistic for which the null hypothesis  is not rejected Critical values : boundaries between the  critical and acceptance regions
Background image of page 2
3 More Definitions Type I error : rejecting the null hypothesis  when it is true Type II error : failing to reject the null  hypothesis when it is false Significance level : probability of type I error Power of the test : probability of (correctly!)  rejecting the null hypothesis when it is false:  (power)=1-(type II error probability)
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4 Hypothesis Testing α  = P(Type I error) = P(reject H 0  | H 0  is true) β  = P(Type II error) = P(accept H 0  | H 0  is false) Power = P(reject H 0 | H 0  is false) = 1 -  β Decision H 0  is true H 0  is false Fail to reject H 0 No error Type II error Reject H 0 Type I error No error
Background image of page 4
5 Statistical inference: sequence of  steps Identify the relevant test statistic (usually  “standardized” version of a good point  estimator) Determine the sampling distribution when the  null hypothesis is true Find the corresponding critical values (from  some table) Get a random sample and conduct the test.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
6 Hypothesis Tests on the Mean H 0 μ  =  μ 0   H 1 μ     μ 0   n X Z 0 0 σ μ - = 2 / 0 2 / 0 2 / 0 2 / 0 0 z Z z if H reject to Fail z Z or z Z if H Reject α - <
Background image of page 6
7 Hypothesis Tests (one side) H 0 μ  =  μ 0   H 1 μ  >  μ 0 H 0 μ  =  μ 0   H 1 μ  <  μ 0 α z Z if H Reject 0 0 z Z if H Reject 0 0 - <
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
8 Example: two-sided test Suppose we take a random sample of  n =25 and  obtains an average horsepower of 143.2. The claimed mean horsepower is 145, and the  standard deviation is known to be 3 hp. We decide to specify a type I error probability  (significance level) of 0.05. Should we doubt the claim?  What would the result be in the one-sided test  case?
Background image of page 8
9 P-value The P-value is the smallest level of significance that  would lead to rejection of the null hypothesis H 0  with the  given data. Another interpretation: the P-value is the probability of 
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 10
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/26/2008 for the course IE 121 taught by Professor Perevalov during the Spring '08 term at Lehigh University .

Page1 / 44

slide4 - Hypothesis testing example An automotive company...

This preview shows document pages 1 - 10. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online