Regression Analysis and Indicator Variables

Abstract

Regression analysis is statistical model that is concerned with describing and evaluating the relationship between a given variable known as dependent variable and one more other variable known as independent variable. The present thesis entitled “Regression Analysis and Indicator Variables”. This exposition comprises four chapters and each chapter is divided into various subsections. Chapter 1 includes introduction about Bivariate distribution. The main focus is on knowing the nature and relationship between two these variables. In this chapter there are two techniques used for this, one is correlation analysis and other is regression analysis. In Chapter 2, simple linear regression is discussed. The main focus in this chapter is on least squares-fit to estimate the model parameters. This chapters includes definition of simple linear regression, examples, properties of least squares estimators and the fitted regression model, estimation of σ 2 , hypothesis testing on model parameters, use of t-tests, testing significance of regression, analysis of variance and coefficient of determination. In Chapter 3, we have discussed multiple linear regression. In this chapter, we have done least squares-estimation of model parameters as described in Chapter 2. This chapters includes basic definition of multiple linear regression, examples, properties of least-square estimators, estimation of σ 2 , testing significance of regression, test on individual regression coefficients and coefficient of determination- R2 and adjusted R2 . Chapter 4 contains introduction about indicator variables. This chapter includes example of indicator variable, an indicator variable with more than two levels, more than one indicator variable.