Notes5 - Lecture Notes 5 Econ 410 Introduction to...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Lecture Notes 5 Econ 410 – Introduction to Econometrics 1 Linear Regression with One Regressor: Part III Heteroskedasticity and Homoskedasticity The error term i u is homoskedastic if the variance of the conditional distribution of i u given i X is constant for n i ,..... , 1 = and in particular it does not depend on i X . Otherwise, the error term is heteroskedastic. Example : consider the following linear regression model: i i i u X Y + + = 0 0 β Assume that i Y is the level of education and i X is a binary variable that equals 0 if the i th person is a woman and equals 1 if the i th person is a man. Let consider the cases 1 = i X and 0 = i X separately. If the person is a woman ( 0 = i X ), the linear regression model can be written as: i i u Y + = 0 and i u is the deviation of the i th woman’s earnings from 0 , the population mean earnings for women. On the other hand, if the person is a man ( 1 = i X ), the linear regression model can be written as: i i u Y + + = 1 0 and i u is the deviation of the i th man’s earnings from 1 0 + , the population mean earnings for men. The error term i u is homoskedastic if the variance of the conditional distribution of i u given i X does not depend on i X . So, in this example, the error term i u is homoskedastic if its variance is the same for men as it is for women, that is: () () 1 | 0 | = = = i i i i X u Var X u Var If this is not the case, then the error term is heteroskedastic. Another example : consider the model in question #4 in Problem Set 2. u inc sav + + = 1 0 , with e inc u = In pa r t b . o f th is ques t ion you show tha t ()
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Lecture Notes 5 Econ 410 – Introduction to Econometrics 2 Implications of Homoskedasticity and Heteroskedasticity We know that, if the three least squares assumptions hold, the OLS estimator is unbiased, consistent and asymptotically normal. These properties hold whether the error term is homoskedastic or heteroskedastic. However, when the error term is homoskedastic, some additional nice properties
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 07/28/2010 for the course ECON 410 taught by Professor Staff during the Fall '08 term at Wisconsin.

Page1 / 5

Notes5 - Lecture Notes 5 Econ 410 Introduction to...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online