Sei sulla pagina 1di 2

STAT 371 Assignment #1 Due Thursday May 17, 2012

1. Five recipes were used for making a number of cakes, starting from two different types A and B of premix. The difference between the two premixes was that A was aerated and B was not. The volumes of the cakes were measured with the following results: Volume recipe 1 2 3 4 5 A 83 90 96 83 90 B 65 82 90 65 82

The five recipes differed somewhat in amount of water added, beating time, baking temp., and baking time. It is claimed that significantly greater volume is obtained with premix A. a) Include in your answer a calculated 95% confidence interval for the true difference between the volumes obtained with A and B. State any assumptions you make. [6 marks] b) Do you think the data support the claim that premix A results in greater volume than premix B? Explain. [2 marks] 2. Fifteen judges rated two randomly allocated brands of beer A and B according to taste (scale 1 to 10) as follows: brand A brand B 2 4 2 1 9 9 2 2 8 3 5 3 7 7 4

a) Stating assumptions, test the null hypothesis that the mean taste scores for A and B are equal. [6 marks] b) Propose a better design for this experiment. Write precise instructions for the conduct of your preferred experiment. [4 marks] 3. Supose we fit a simple linear regression model of the form: Yi = 0 + 1 ( xi x ) + Ri , where

Ri ~ G ( o, ) , i = 1, 2, , n independent and x is the average


a) Show that

x
i

n.

i =1 i

r = 0 where ri is the estimated residual that results from the fit of a simple linear

regression model using the least squares solution. [4 marks] n b) Also show that i =1 xi ri = 0 . [4 marks] c) Distinguish and explain the difference between i) residual and standardized residual [2 marks] n ii) E ( Ri ) = 0 and i =1 ri n = 0 [2 marks]

4. KW Office Equipment Corporation sells an imported photo copy machine on a franchise basis and performs preventive maintenance and repair service on this machine. The data below have been collected from 18 recent service calls; for each call, x is the number of machines serviced and y is the total number of minutes spent by the service person. Assume a simple linear regression model of the form Yi = 0 + 1 ( xi x ) + Ri , where Ri ~ G ( o, ) , i = 1, 2, , n independent and x is the average

x
i

n , is appropriate. 2 6 86 3 5 78 4 1 10 5 5 75 6 4 62 7 7 101 8 3 39 9 4 53 10 2 33 11 8 118 12 5 65 13 2 25


2

i xi yi

1 7 97

14 5

15 7

16 1

17 4 49

18 5 68

71 105 17
= , 16,504

Summary calculation results are:

(x x )
i

=, 74.5

1098 ( y y )( x x ) = .
i i

= 1152 ,

= 81 ,

( y y )
i

Use hand calculations for parts a)-g)

a) Obtain the estimated (fitted) regression equation. [2 marks] b) Plot the estimated regression function and the data. Also plot the residuals vs. the explanatory variable (x). How well does the estimated regression equation fit the data? [4 marks] c) Interpret 0 in your regression equation. Does 0 provide any relevant information here? Explain. [4 marks] d) Obtain a 95% confidence interval for 1 . [6 marks] e) Conduct a t-test to determine whether or not there is a linear association between x and y. State the alternative hypothesis, decision rule, and the conclusion. What is the p-value of your test? [4 marks] f) Obtain an approximate 95% prediction interval for the service time on the next service call in which six machines are serviced. Interpret your prediction interval. [6 marks]

Potrebbero piacerti anche