LOYOLA COLLEGE (AUTONOMOUS), CHENNAI – 600 034
M.Sc. DEGREE EXAMINATION – STATISTICS
FOURTH SEMESTER – APRIL 2011
ST 4809 / 4805 – APPLIED EXPERIMENTAL DESIGN
Date : 5/4/2011 Dept. No. Max. : 100 Marks
Time : 9:00 – 12:00
SECTION – A
Answer all the questions (10 x 2 = 20 marks)
- State the linear model used in 23 factorial design
- Briefly explain the term Blocking.
- When do we go for repeated L.S.D ?
- State Fermat’s theorem.
- Define the term linear contrast with an example.
- State the missing formula for RBD.
- Define a resolvable BIBD with an example .
- Distinguish between RBD and BIBD.
- Give any two advantages of factorial design.
- State any two applications of response surface design.
SECTION-B
Answer any Five questions (5 x 8 = 40 marks)
- Distinguish between CRD and RBD and derive the efficiency of RBD over CRD.
- Describe the analysis of variance for a GRAECO-LATIN SQUARE, stating all the
Hypothesis, ANOVA and conclusions.
- Define mutually orthogonal contrast and show that in 23 design the
main effects and interaction effects are mutually orthogonal.
- Discuss in detail the need for a Nested design with suitable illustration.
- Describe, the analysis of variance for a 33 factorial design, stating all the
hypothesis, ANOVA and conclusions.
- Develop the analysis of variance for a 24 completely confounded factorial design.
- State clearly the model used in the case of Youden Square and construct a real life
example.
- Describe in detail the LINEAR and QUADRATIC response surface Designs.
SECTION-C
Answer any Two questions (2 x 20 = 40 marks)
- Discuss in detail the real life scenario of application of experimental designs in
- i) Agriculture ii) Industry iii) Health sciences. ( 6+7+7 marks)
20.a) Define the term key BLOCK in the case of 25 factorial design of size 23 in
which 2 independent interactions and 1 generalized interaction are Confounded
Discuss in detail using the required linear equations for the confounded effects.
- b) State and prove any one of the parametric conditions of a BIBD
(15 +5-Marks)
21 a) When do we go for PBIBD?
- b) Construct a PBIBD with 2 associate classes stating all the parametric conditions
- c) Explain the m-ple Lattice Square design and hence construct lattice square design
when the block size k=3. (2+8+10-Marks)
22 Write short notes on the following
- Replication
- MOLS
- Partial Confounding
- Split plot (5+5+5+5-Marks)