1 / 83

840 likes | 1.11k Vues

Econ 240A. Power 6. The Challenger Disaster. http://onlineethics.org/moral/boisjoly/RB-intro.html. The Challenger. The issue is whether o-ring failure on prior 24 prior launches is temperature dependent They were considering launching Challenger at about 32 degrees

Télécharger la présentation
## Econ 240A

**An Image/Link below is provided (as is) to download presentation**
Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.
Content is provided to you AS IS for your information and personal use only.
Download presentation by click this link.
While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.
During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

**Econ 240A**Power 6**The Challenger Disaster**• http://onlineethics.org/moral/boisjoly/RB-intro.html**The Challenger**• The issue is whether o-ring failure on prior 24 prior launches is temperature dependent • They were considering launching Challenger at about 32 degrees • What were the temperatures of prior launches?**Only 4 launches**Between 50 and 64 degrees Challenger Launch**Challenger**• Divide the data into two groups • 12 low temperature launches, 53-70 degrees • 12 high temperature launches, 70-81 degrees**Probability of O-Ring Failure Conditional On Temperature,**P/T • P/T=#of Yeses/# of Launches at low temperature • P/T=#of O-Ring Failures/# of Launches at low temperature • Pˆ = k(low)/n(low) = 5/12 = 0.41 • P/T=#of Yeses/# of Launches at high temperature • Pˆ = k(high)/n(high) = 2/12 = 0.17**Are these two rates significantly different?**• Dispersion: p*(1-p)/n • Low: [p*(1-p)/n]1/2 = [0.41*0.59/12]1/2 =0.14 • High: [p*(1-p)/n]1/2 = [0.17*0.83/12]1/2 =0.11 • So .41 - .17 = .24 is 1.7 to 2.2 standard deviations apart? Is that enough to be statistically significant?**Outline**• Interval Estimation • Hypothesis Testing • Decision Theory**The Field Poll, 10-7-’04**• In a sample of approximately 1135 likely voters, 48% indicate they will vote for Senator Boxer Vs. 32% for Jones • If the poll is an accurate reflection or subset of the population of voters Nov. 2, what is the expected proportion that will vote for Boxer? • How much uncertainty is in that expectation? Power 4**Power 4**Field Poll • The estimated proportion, from the sample, that will vote for recall is: • where is 0.48 or 48% • k is the number of “successes”, the number of people sampled who are for Boxer, approximately 545 • n is the size of the sample, 1135**Field Poll**Power 4 • What is the expected proportion of voters Nov. 2 that will vote for Boxer? • = E(k)/n = np/n = p, where from the binomial distribution, E(k) = np • So if the sample is representative of voters and their preferences, 48% should vote for Boxer in three weeks.**Power 4**Field Poll • How much dispersion is in this estimate, i.e. as reported in newspapers, what is the margin of sampling error? • The margin of sampling error is calculated as the standard deviation or square root of the variance in • = VAR(k)/n2 = np(1-p)/n2 =p(1-p)/n • and using 0.48 as an estimate of p, • = 0.48*0.52/1135 =0.00022**Interval Estimation**• Based on the Poll of 48% for Boxer, what was the probability that the fraction, p, voting for Boxer would exceed 50%, i.e. lie between 0.5 and 1.0? • The standardized normal variate, z =**Interval estimation**• Why can we use the normal distribution? • Where does the formula for z come from?**Solving for p:**.015*z = 0.48 - p p = 0.48 -.015*z and substituting for p: and subtracting 0.48 from each of the 3 parts of this inequality:**And dividing by –0.015,**which changes the signs of the inequality: And using the standardized normal distribution, this probability equals ….0.5**0**-34.7**0**-34.7**Solving for p:**.015*z = 0.48 - p p = 0.48 -.015*z and substituting for p: and subtracting 0.48 from each of the 3 parts of this inequality:**And dividing by –0.015,**which changes the signs of the inequality: And using the standardized normal distribution, this probability equals ….0.092**-1.33**-34.7**-1.33**-34.7**So a Z value**of 1.33 leads to an area of 0.408, leaving 0.092 in the Upper tail**Interval Estimation**• The conventional approach is to choose a probability for the interval such as 95% or 99%**So z values**of -1.96 and 1.96 leave 2.5% in each tail**1.96**-1.96 2.5% 2.5%**Substituting for z**And multiplying all three parts of the inequality by 0.015**And subtracting 0.48 from all three parts of the inequality**And multiplying by -1, which changes the signs of the inequality: So a 95% confidence interval based on the poll, predicted a vote for Boxer of between 45% and 51%, an inference about the unknown parameter p. Z values of -2.575 and 2.575 leave 1/2% in each tail. You might calculate a 99% confidence interval for the poll.**Two Californias**http://www.sfgate.com/election/races/2003/10/07/map.shtml**Interval Estimation**• Sample mean example: Monthly Rate of Return, UC Stock Index Fund, Sept. 1995 - Aug. 2004 • number of observations: 108 • sample mean: 0.842 • sample standard deviation: 4.29 • Student’s t-statistic • degrees of freedom: 107**Sample**Mean 0.842**Appendix B**Table 4 p. B-9 2.5 % in the upper tail**Interval Estimation**• 95% confidence interval • substituting for t**Interval Estimation**• Multiplying all 3 parts of the inequality by 0.413 • subtracting .842 from all 3 parts of the inequality,**Interval EstimationAn Inference about E(r)**• And multiplying all 3 parts of the inequality by -1, which changes the sign of the inequality • So, the population annual rate of return on the UC Stock index lies between 19.9% and 0.2% with probability 0.95, assuming this rate is not time varying**Hypothesis Testing: 4 Steps**• Formulate all the hypotheses • Identify a test statistic • If the null hypothesis were true, what is the probability of getting a test statistic this large? • Compare this probability to a chosen critical level of significance, e.g. 5%**Hypothesis Test Example**• Field Poll on Boxer • Step #1: null, i.e. the maintained, hypothesis: true proportion for Boxer is 50% H0 : p = 0.5; the alternative hypothesis is that the true population proportion supporting Boxer is greater than 50%, Ha : p>0.5**Hypothesis Test Example**• Step #2: test statistic: standardized normal variate z • Step #3: Critical level for rejecting the null hypothesis: e.g. 5% in upper tail; alternative 1% in upper tail**Step #4: compare the**probability for the test statistic(z= -1.33) to the chosen critical level(z=1.645) Sample statistic 1.645 5 % upper tail 6**Hypothesis Test Example**• So, since –1.33 is not above the critical value of 1.645, I.e. not extreme and not in the upper tail, do not reject the null hypothesis that p=0.5 . • In terms of common sense, the sample proportion of 0.48 means p=0.5 is more likely than the alternative of p>0.5.**Also recall the 95% confidence interval on p which was**between 0.45 and 0.51, including the null that p= 0.5.**And subtracting 0.48 from all three parts of the inequality**And multiplying by -1, which changes the signs of the inequality: So a 95% confidence interval based on the poll, predicted a vote for Boxer of between 45% and 51%, an inference about the unknown parameter p. Z values of -2.575 and 2.575 leave 1/2% in each tail. You might calculate a 99% confidence interval for the poll.

More Related