YB 28

    LOYOLA COLLEGE (AUTONOMOUS), CHENNAI – 600 034

B.Sc. DEGREE EXAMINATION – STATISTICS

SIXTH SEMESTER – April 2009

ST 6600 – DESIGN & ANALYSIS OF EXPERIMENTS

 

 

 

Date & Time: 18/04/2009 / 9:00 – 12:00             Dept. No.                                                   Max. : 100 Marks

 

 

PART – A

 

Answer ALL the questions.                                                                               (10 X 2 = 20 marks)

 

1. Give a contrast orthogonal to the contrast
2. What is a mixed effect model?
3. Mention the principle of experimental design which is not taken care by CRD if fertility gradient is present in one direction
4. Give the skeleton ANOVA table for two-way classification with interaction
5. What should be the number of treatments if the error degrees of freedom associated with a Randomised Block Design with 5 blocks is 12?
6. What are Latin Squares ? Give an example of a latin square.
7. Write all possible treatment combinations in a design.
8. When do you recommend confounding ?
9. Write the factorial effect corresponding to in design
10. If the interactions BCD and ABE are confounded in a replication consisting of four blocks of 8 plots in factorial design which effect also gets confounded automatically.

 

PART – B

 

Answer any FIVE Questions.                                                                            (5 X 8 = 40 marks)

     

1. Explain in detail the principles of experimental design.
2. Write a descriptive note which highlights the difference between fixed effect and random effect models.
3. Develop ANOVA for one way classified data
4. Derive the formula for estimating the missing value in a LSD when one observation is missing.
5. Explain the preparation of layout of a 5 x 5 Latin square design.
6. Explain the process of computing various factorial effects in the case of a design.
7. Illustrate with an example the method of finding the contrast defining the interaction between a Quadratic Effect and Linear Effect in the case of a factorial design involving two factors with three levels each.
8. It is decided to confound an effect in the case of a experimental design.

• What will be your choice of the effect to be considered for confounding ? Justify your answer.
• Explain the preparation of lay out for such a confounding in 4 replications.

 

PART – C

 

Answer any TWO Questions.                                                                            (2 x 20 =40 marks)

 

1. Describe in detail the preparation of layout for a randomized block design and the  steps involved in its analysis.
2. Describe fully the analysis of a factorial design
3. Develop the complete analysis of Balanced Incomplete Block Design
4. Write short notes on the following:

• Efficiency of RBD over CRD b) Cochran’s theorem   c) Generalized Effect

(d) Partial Confounding

 

 

Go To Main Page

Latest Govt Job & Exam Updates:

View Full List ...