This textbook makes learning the basic principles of econometrics easy for all undergraduate and graduate students of economics. It specifically caters to the syllabus of the Introductory Econometrics course taught in the third year of the Bachelor of Economics program in many universities.

It takes the readers step-by-step from introduction to understanding, first introducing the basic statistical tools like concepts of probability, statistical distributions, and hypothesis tests, and then going on to explain the two variable linear regression models along with certain additional tools like use of dummy variables, various data transformations amongst others.

The most innovative feature of this textbook is that it familiarizes students with the role of R, which is a flexible and popular programming language. With its help, the student will be able to implement a linear regression model and deal with the associated problems with substantial confidence.

# Heteroskedasticity, Autocorrelation and Issues of Specification

### Heteroskedasticity, Autocorrelation and Issues of Specification

Heteroskedasticity, autocorrelation and issues of specification

### Introduction

In Chapters 4 and 5, we studied the Ordinary Least Squares (OLS) model. We learnt how to calculate the OLS coefficients and also what their small sample and asymptotic properties are. In particular, we proved that if we are willing to make appropriate assumptions on the stochastic error term as well as the set of explanatory variables, the OLS estimators are unbiased, efficient and consistent, apart from being normally distributed. More formally, we argued that: We also showed that the variance of any other linear unbiased estimator will be at least as ‘large’ as the variance of the OLS estimator.

There is no reason why all the assumptions that we have made on the stochastic error ...

• • • 