Course Contents

Logic of Regression and correlation, scatter diagram, simple linear regression model, least square estimators of parameters, standard error of estimate. And inference about regression coefficients. Multiple linear regression with two repressors. Simple, multiple and partial correlation up to three variables. Coefficient of multiple determination. Meaning, derivation and application of rank correlation, tied ranks. Inference about simple population correlation coefficient. Concept of analysis of variance. Testing the equality of several means by analysis of variances for one-way classification (equal and unequal number of observations) and two-way classifications. The LSD test. Basic principles of experimental design, Completely Randomized, Randomized Complete Block and Latin Square Designs: their descriptions, layout, statistical analysis, advantages and disadvantages. Introduction to nonparametric statistics, usage of nonparametric statistical methods, Concept and use of one or two sample nonparametric testing and estimation methods. The sign test, the wilcoxon rank sum test, the mann-whitney u test, the median test, the runs test, the kolmogrov-smirnov test, the kruskal-wallis test, friedman test for randomness.. Case Study: Analyze the real data and present the work in the form of report.

Course Synopsis

Introduction to data handling techniques, testing of hypothesis and non parametric tests.

Course Learning Outcomes

1. Students should be able to handle linear and multiple linear modeling on real life data 2. They can easily understand how to form structure of dependence through modeling 3. Students would be able to handle multiple treatments or can check the efficiency of multiple treatments for experimental data