This preview shows pages 1–9. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Click to edit Master subtitle style ECON1203/ECON2292 Business and Economic Statistics Week 9 22 Week 9 topics l Calculating probability of type 2 errors l Power of test l Hypothesis testing & confidence intervals for the mean when population variance is unknown l t distribution l Sampling distribution of sampling proportion l Hypothesis testing & confidence intervals for the population proportion l Key references l Keller 11.311.4, 12.1, 9.2 pp 30914, 12.3 33 Type I errors in hypothesis testing l Type I errors occur when we reject a true null hypothesis l Denote P (Type I error) = P (Reject H 0 H 0 true) = l Analyst specifies l Typical values chosen for in applications are 0.10, 0.05 or 0.01 l Value of is called the significance level of the test Type II errors in hypothesis testing l Type II errors occur when we dont reject a false null hypothesis l Denote P (Type II error) = l For given there will be a range of values (depending on the true alternative) l Can be calculated for all values under alternative P (Do not reject H 0  H 0 not correct) = l Changing for given fixed alternative changes l There is a tradeoff between the 2 types of errors 44 55 Calculating probability of Type II errors l Recall McDonalds example l H 0: = 0.25, H 1: < 0.25 l With n = 25, = 0.05, = 0.05 previously found a decision rule (DR) of rejecting H 0 if sample mean < 0.2336 l Suppose that McDonalds only puts 0.24 pounds in their quarter pounder. Will we detect this? 66 Power of a test l Defined as the probability of correctly rejecting a false null hypothesis l P (Do not reject H 0 H 0 not correct) = l P (Reject H 0 H 0 not correct) = Power = 1 l Keller p. 370, (Fig. 11.10) l H 0: = 170; H 1: > 170 l n = 400, = 0.01, = 65 l DR: reject H 0 if sample mean > 177.57 l Power if = 180? 77 Power of a test l What happens to power if increase to 0.05? l What happens to power if increase n to 1000? Power of a test... l Summary l H 0: = 170; H 1: = 180 l If = 0.01, then = 0.2266 ( Power = 0.7734) when n= 400 l If = 0.05, then = 0.0764 ( Power = 0.9236) when n= 400 l If = 0.05, then 0.0000 ( Power 1) when n= 1000 l With a different alternative e.g. With a different alternative e....
View
Full
Document
This note was uploaded on 07/28/2011 for the course ECON 1203 taught by Professor Denzilgfiebig during the Three '11 term at University of New South Wales.
 Three '11
 DenzilGFiebig

Click to edit the document details