Sample Past Paper CVEN2002
Sample Past Paper CVEN2002
Sample Past Paper CVEN2002
THE UNIVERSITY OF NEW SOUTH WALES SCHOOL OF MATHEMATICS AND STATISTICS November 2011
(1) TIME ALLOWED 3 Hours (2) TOTAL NUMBER OF QUESTIONS 3 (3) ANSWER ALL QUESTIONS (4) THE QUESTIONS ARE OF EQUAL VALUE (5) THIS PAPER MAY NOT BE RETAINED BY THE CANDIDATE (6) ONLY CALCULATORS WITH AN AFFIXED UNSW APPROVED STICKER MAY BE USED (7) STATISTICAL FORMULAE ARE AT THE START OF PART B STATISTICAL TABLES ARE ATTACHED AT END OF PAPER
Part A consists of questions 1 3 Part B Statistics consists of questions 4 6 Both parts must be answered
All answers must be written in ink. Except where they are expressly required pencils may only be used for drawing, sketching or graphical work.
November 2011
Page 2
Part B Statistics
1.
b) A k -out-of-n system is one that will function if and only if at least k out of the n individual components in the system function. Assume that individual components function independently of each other. Assume also each individual component functions with probability 0.9.
November 2011 2.
Page 3
i) Calculate the value represented by *. What does this number represent in the study? ii) Set up suitable notation, and state the null and alternative hypotheses for the ANOVA F-test. What assumptions need to be made for an Analysis of Variance to be an appropriate analysis here? iii) Using a signicance level of = 0.01, (and making the necessary assumptions) complete out the ANOVA F-test. Include in your answer: the observed value and distribution of the test statistic; a mathematical expression for the p-value; the range of values within which the p-value falls (and a statement of how these are obtained); and the conclusion of the test stated in plain language. b) An article in The Engineer reported the results of an investigation into wiring errors on commercial transport aircraft that may produce faulty information to the ight crew. Of 1600 randomly selected aircraft, eight were found to have wiring errors that could display incorrect information to the ight crew. i) Find an approximate 99% two-sided condence interval on the proportion of aircraft that have such wiring errors. ii) How large a sample would be required if we wanted to be at least 99% condent that the observed sample proportion p diers from the true proportion by at most 0.008, regardless of the value p ? c) Suppose that X and Y are independent standard normal variables: X N (0, 1), i) What is the distribution of X + Y ? ii) Calculate P (X + Y < 1). Please see over . . . Y N (0, 1).
November 2011 3.
Page 4
Using the regression analysis output (which includes residual plots and a tted line plot) at the end of the question, answer the following questions. a) i) Write the null and alternative hypotheses to test whether the variable ppv is signicant in predicting the variable Ratio. ii) Carry out the test at the 1% signicance level. i) Determine the value of R2 ; ii) hence determine the (sample) correlation between the variables Ratio and ppv. i) Determine a 95% condence interval for 1 ; ii) hence determine a 95% condence interval for the change in the mean of Ratio for an increase of 100 mm/sec in ppv.
b)
c)
d) In the raw data, what (approximately) were the range of values for the peak particle velocity? e) For the regression analysis to be valid, the error in Model ( ) must be independent, and normally distributed N (0, ) for some constant . This part of this question is about the assumption that the standard deviation is constant: i) There are four residual plots. Which of these can be used to verify whether the constant assumption is plausible? Explain whether this assumptions is supported, with specic reference to the regression analysis output. ii) What is the estimate of from the regression analysis of the data? f) Determine a 95% condence interval for 0 .
November 2011
Page 5
T 761.91 -7.80
P 0.000 0.000
S = 0.00314906
R-Sq = 68.5%
R-Sq(adj) = 67.4%
Analysis of Variance Source Regression Residual Error Total DF 1 28 29 SS 0.00060380 0.00027766 0.00088147 MS 0.00060380 0.00000992 F 60.89 P 0.000
Unusual Observations Obs 10 ppv 486 Ratio 0.985000 Fit 0.992864 SE Fit 0.000629 Residual -0.007864 St Resid -2.55R
R denotes an observation with a large standardized residual. Predicted Values for New Observations New Obs 1
Fit 0.988947
SE Fit 0.000625
ppv 750
Note: Regression analysis continues on the next page with residual plots and a tted line plot. Please see over . . .
November 2011
MATHXXXX
Page 6