Description
Random effect and mixed effect have an active area of research in statistics. A random effects model is an analysis of variance (ANOVA) model where all the effect are random. If the model contains both fixed effects and random effects, it is referred to as a mixed effect model. The variance of random effects are called variance component models. Such models are extensively used for the analysis of agricultural, industrial and genetic data . In practical applications involving a variance component and testing the significance of the fixed effects and variance components.
In the short course, mixed and random effects models will be described and illustrated using a number of practical examples. Methods of point estimation, confidence intervals and hypothesis testing will be described and explained ,once again , using practical examples . The difficulties associated with unabalanced data analysis will be highlighted and available procedures will be described . Even though mixed and random models have a long history, and the literature is fairly extensive, many significant developments took place only recently, especially in the context of hypothesis testing and confidence intervals. In fact, satisfactory test procedures for many problems have been derived only recently ( during the 1990's); for example, the derivation of hypotheses tests based on combined information ( recovery of inter-block information ),and the development of hypotheses tests and confidence intervals based on generalized p-values and generalized confidence intervals, to handle some non standard problems that are encountered in practice. The short course will introduce and discuss the traditional as well as the latest methodologies for data analysis based knowledge on mixed and random models.
The anticipated audience is expected to have basic knowledge of regression and ANOVA.
|
