Ass1 mth513

Download as doc, pdf, or txt
Download as doc, pdf, or txt
You are on page 1of 3

Assignment No.

1
MTH 513A-Analysis of Variance

1. The response time in milliseconds for 5 different type of circuits in automatic


value shutoff mechanism is measured and reported as follows

Circuit 1 Circuit 2 Circuits 3 Circuits 4 Circuit 5


28 25 31 31 29
25 31 25 25 31
28 30 31 25 23
25 22 23 31 28
29 27 38 21 24
35 29 29 31 19
31 29 28 30 25
23 29 23 27
35 34 22 25
24 22 31
29 25
39
Test the hypothesis that all 5 circuits have same response time. Use a = .05 . Use
further critical difference and Scheffe’s methods for multiple comparison, if
needed.

2. Suppose a pharmaceutical research company conducts an experiment to compare


the efficiency of 5 drugs. These are 20 animals available for the trial and each
drug is injected into 4 randomly selected animals. 3 animals die during the course
of the experiment. The blood samples from the remaining animals are taken and
analyzed. The data on blood pH reading from each blood analysis in certain
standardized units are given in the following table.
Drugs
_____________________________________________________________
A B C D E

19 7 4 6 6
11 1 7 6 4
15 4 7 6 2
4 10
_____________________________________________________________

i) Analyse the data and test if all the drugs are equally effective or not.
ii) Consider the contrast
L = m1 + 2m 2 - m3 - m4 - m5
Test H 0 : L = 0 versus H1 : L �0.
iii) Apply critical difference method and Scheffe’s procedure for multiple
comparison test, if needed.
3. The tensile strength of cement is studied. The data on 4 different methods of
mixing is collected as follows:
Mixing Technique Tensile Strength ( lb / in 2 )
________________________________________________
1 3129 3000 2865 2890
2 3200 3300 2975 3150
3 2800 2900 2985 3050
4 2600 2700 2600 2765
_______________________________________________
(a) Test the hypothesis that mixing techniques affect the strength of the cement.
Use a = 0.05.
(b) Use Student Newman Keul’s test, Duncan’s test and Tukey’s procedure to
make comparisons between pairs of means, if needed.

4. An engineer is interested in investigating the tensile strength of a synthetic


fibre. The engineer knows from his experience that the tensile strength is
affected by the weight percent of cotton used in the blend of materials for
the fibre. He decides to test specimens at 5 levels and to test five
specifications at each level of content. He obtains the following data on
tensile strength (lb / in 2 )

Cotton weight Observations


Percentage
1 2 3 4 5
15 7 7 15 11 9
20 12 17 12 18 18
25 14 18 18 19 19
30 19 25 22 19 23
35 7 10 11 15 11
(i) Use thr model ijy = m i + eij with usual notations and test of cotton weight

percentage in synthetic fibre affects the tensile strength at 5% level of


significance. Interpret the resuls.
(ii) In case, the null hypotesis (equality of means) is rejected, use Duncan’s
test, Student-Newmann Keuls test, Scheffe’s and Tukey’s test to find that
(iii) which pairs of treatment are responsible for the rejection of null
hypothesis.
(iv) Suppose the engineer wants to test the following null hypothesis
H 01 : m 4 = m 5
H 02 : m1 + m 3 = m 4 + m 5
H 03 : m1 = m 3
H 04 : 4 m 2 = m1 + m 3 + m 4 + m5

Prepare ANOVA table to test these hypothesis and give your conclusion.
5 A scientist wishes to test the effect of four types of sleeping
medicines. The experiment is designed with 24 patients of the same
physical and clinical characteristics. The administers each type of
medicine at random in 6 of the 24 patients. The response (in some
appropriate units) of 24 patients is given below.
A B C D
96 93 60 76
37 81 54 89
58 79 78 88
69 101 56 84
73 96 61 75
81 102 69 68
Consider a fixed effect model yij = m + a i +ij with the usual assumptions
and rotations.
(i) Conduct a test of hyp H o :a1 = a 2 = a 3 = a 4. and construct the
ANOVA table .
(ii) Apply t-test, student –Newman-Keuls test, Duncan’s test, Tukey’s test
and Scheffe’s test in case H 0 is rejected.
(iii) Consider a contrast
L =a1 + a 2 - a 3 - a 4.
Test H o : L = 0 versus H 1 : L1 0
Construct the upper and lower bounds for L in (iii) using Tukey’s and Scheff’s
procdure.

You might also like