Stat 344 Lecture 20 (20, 21)

Stat 344 Lecture 20 (20, 21) - Confidence Intervals Lecture...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Confidence Intervals Lecture 20 Topics Variances Proportions Prediction intervals Lecture 19 R eference: Montgomery Sections 9-1 thru 9-3 Devore Lecture 20 Devore Lecture 21
Background image of page 1

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

View Full DocumentRight Arrow Icon
Hypothesis and Test Procedures Lecture 20 Topics Hypothesis tests versus confidence intervals. How to write hypothesis statements Type I and Type II errors & their probabilities. Concept of the critical cutoff statistic & the rejection region. Defined alternative hypotheses Lecture 20 Reference: Devore 2 Stat 344 Lecture 20
Background image of page 2
What are Hypotheses Hypothesis: A claim or assertion about a population parameter. H 0: is the symbol for the null hypothesis, The current value, The status quo. H a: is the symbol for the alternative hypothesis, aka H 1:, regards a claim. Stat 344 Lecture 20 3
Background image of page 3

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

View Full DocumentRight Arrow Icon
Hypothesis Structure H 0: μ = 90, 90 is the hypothesized value of μ. Parameter could be σ2 or p. H a: μ ≠ 90 < for a left-tail test > for a right-tail test ≠ for a two-tail test We seek to reject the null hypothesis if there is sufficient evidence to do so. Stat 344 Lecture 20 4
Background image of page 4
Hypothesis Decision Matrix α = Prob(Type I error). Type I error is the error most to be avoided. β = Prob(Type II error). Stat 344 Lecture 20 5 Our decision H0 is true μ = 90 Ha is true μ ≠ 90 Do not reject H 0 OK Type II error Reject H 0 Type I error OK Reality
Background image of page 5

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

View Full DocumentRight Arrow Icon
Example 8.1 Bumper Damage-a An car bumper sustains no visible damage 25% of the time in a crash test. A proposed redesign seeks to increase that percentage. The redesign will be tested with a sample of n =20 crashes. H 0: p = 0.25, H a: p > 0.25 Stat 344 Lecture 20 6 X is the number of crashes with no visible damage, a binomial rv. Rejection region R 8 = {8, 9, 10,…,19, 20} when x ≥ 8 and H 0: p = 0.25. α = P (Type I error) = P ( H 0 rejected when true) = P ( x ≥ 8) = 1– P ( x ≤ 7) = 0.102 x P(X ≤ x) 0 0.003 1 0.024 2 0.091 3 0.225 4 0.415 5 0.617 6 0.786 7 0.898 8 0.959 9 0.986 10 0.996 20 1.000 B(n=20, p=0.25 0.20 0.15 0.10 0.05 0.00 X Probability 8 0.102 0 Binomial, n= 20, p= 0.25
Background image of page 6
Example 8.1 Bumper Damage-b Now suppose H a: p =0.3 β(0.3) = P (Type II error when p = 0.3) = P ( H 0 is not rejected when it is false because p = 0.3) = P(x ≤ 7) = B(7, 20, 0.3) = 0.772 Stat 344 Lecture 20 7 0.20 0.15 0.10 0.05 0.00 X Probability 7 0.132 17 Binomial, n= 20, p= 0.5 p β(p) 0.3 0.772 0.4 0.416 0.5 0.132 0.6 0.021 0.7 0.001 0.8 0.000
Background image of page 7

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

View Full DocumentRight Arrow Icon
Example 8.2 Paint Drying Time-a The drying time of a paint is known to be a normally distributed rv with μ = 75 min and σ = 9 min . An additive is proposed to decrease the drying time. Because of the high cost of the additive, there should be strong evidence that it decreases the drying time. Test will be based on 25 samples of paint with the additive. Stat 344 Lecture 20
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/20/2011 for the course STAT 344 taught by Professor Staff during the Spring '08 term at George Mason.

Page1 / 39

Stat 344 Lecture 20 (20, 21) - Confidence Intervals Lecture...

This preview shows document pages 1 - 9. Sign up to view the full document.

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