Design of Experiments
Design of Experiments provides the statistical tools to get maximum information from least amount of resources. This course is meant to expose the students to the basic principles of design of experiments. The students would also be provided with mathematical background of various basic designs involving one-way and two way elimination of heterogeneity and their characterization properties. This course would also prepare the students in deriving the expressions for analysis of experimental data.
- Basic principles of design of experiments: Randomization, Replication and local control. Uniformity trials: Shape and size of plots and blocks. Elements of linear estimation. Analysis of variance and covariance.
- Completely randomized, Randomized block and Latin square designs.
- Factorial experiments, Confounding in symmetrical factorial experiments (2n and 3n series), Split plot and Strip-plot designs. Mutually orthogonal latin squares. Missing plot techniques. Graeco Latin squares.
- Balanced incomplete block (BIB) designs: General properties and analysis with and without recovery of information. Construction of BIB designs, Youden square designs, Lattice designs. Change-over designs. Groups of experiments.
- Bioassays: Direct and indirect, Indirect assays based on quantal dose response, Parallel line and slope ratio assays potency estimation.
Uniformity trial data analysis, Formation of plots and blocks, Fairfield Smith Law. Analysis of data obtained from CRD, RBD, LSD. Analysis of factorial experiments without and with confounding. Analysis with missing data. Split plot and strip plot designs. Transformation of data. Analysis of resolvable designs. Fitting of response surfaces.