The main objective of this course is to provide a strong mathematical and conceptual foundation in the methods of Bayesian analysis, with an emphasis on practical aspects of the interpretation and communication of statistically based conclusions in research. Bayesian procedures are concerned with best estimating a value or range of values for a particular population parameter. It deals with the estimation of parameters in different approach.Bayesian analysis is a system for describing epistemological uncertainty using the mathematical language of probability. Bayesian analysis, a method of statisticalinference  that allows one to combine prior information about a population parameter with evidencefrom information contained in a sample to guide the statistical inference process.

Contents

Bayesian Statistics, Formulation of a decision problem: randomized and non-randomized decision rules, risk function, optimality of decision rules. Utility theory and loss function. Subjective probability and selection of prior distribution for Bayesian analysis. Bayesian analysis for statistical inference problems of estimation, testing hypotheses, confidence interval and prediction. Bayesian decision theory. Admissible and minimax decision rules. Complete class of decision rules.

Books Recommended:

1.         De Groot M. H. (2004). Optimal Statistical Decisions. Wiley

2.         James O. Berger (1985). Statistical Decision Theory and Bayesian Analysis. Second Edn. Springer -Verlag

Distribution of Marks:

Mid Exam:           30
Final exam:         50
Sessional (Assignment,Presentation,Participation,Attendance,Quizes)    20

Scheduled on:      
M.Phil:        Thursday(2:00-5:00)