# However without assumption iv we need an additional

• Notes
• 29

This preview shows pages 17–20. Sign up to view the full content.

However, without Assumption IV * we need an additional condition in Proposition 6 in order to use the central limit theorem, namely: Proposition 10 . If the sample size n is large then under the assumptions I * - III * and the additional condition the approximate normality results in Proposition 7 carry over. E [ X 2 j ] < 4 Moreover, without Assumption IV * the Propositions 6 and 8 are no longer true. As to Proposition 6, this not a big deal, as in large samples we can still use Proposition 7, but without Assumption IV * we can no longer derive confidence intervals for the forecasts, as these confidence intervals are based on Proposition 8. It is therefore important to test the normality assumption. 10. Testing the normality assumption For a normal random variable U with zero expectation and variance σ 2 it can be shown that

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

3 Jarque, C.M.and A.K. Bera, (1980), "Efficient Tests for Normality, Homoscedasticity and Serial Independence of Regression Residuals". Economics Letters 6, 255 BB 259. 4 Kiefer, N. and M. Salmon (1983), "Testing Normality in Econometric Models", Economic Letters 11, 123-127. 5 Also spelled as "Heteroskedasticity." 18 Kurtosis ' def . E [ U 4 ]/ F 4 & 3 ' 0, Skewness ' def . E [ U 3 ] ' 0 (34) Therefore, the normality condition can be tested by testing whether the kurtosis and the skewness of the model errors are zero, using the residuals. This is the idea behind the Jarque-Bera 3 and Kiefer-Salmon 4 tests. Under the null hypothesis (34) the test statistic involved has a χ 2 2 distribution 11. Heteroscedasticity 5 We say that the errors U j of regression model (2) are heteroskedastic if assumption III * does not hold: E [ U 2 j | X j ] ' R ( X j ) for some function R (.). (35) Heteroscedasticity often occurs in practice. It is actually the rule rather than the exception. The main consequence of heteroscedasticity is that the conditional variance formulas in Propositions 2 and 3 do no longer hold, although the unbiasedness result in Proposition 1 is not affected by heteroscedasticity. Therefore, the Propositions 4-8 are no longer valid as well. In particular, the conditional variance of [see (60)] under heteroscedasticity takes the form ˆ β var( \$ \$ | X 1 ,..., X n ) ' E [( \$ \$ & \$ ) 2 | X 1 ,..., X n ] ' ' n j ' 1 ( X j & ¯ X ) 2 R ( X j ) ' n i ' 1 ( X i & ¯ X ) 2 2 . (36) A cure for the heteroscedasticity problem is to replace the standard error of by ˆ β
6 Breusch, T. and A. Pagan (1979), "A Simple Test for Heteroscedasticity and Random Coefficient Variation", Econometrica 47, 1287-1294. 7 In the multiple regression case the degrees of freedom is equal to the number of parameters minus 1 for the intercept. 19 # F \$ \$ ' n n & 2 ' n j ' 1 ( X j & ¯ X ) 2 \$ U 2 j ' n i ' 1 ( X i & ¯ X ) 2 2 . (37) This is known as the Heteroscedasticity Consistent (H.C.) standard error. The H.C. t-value then becomes Under the null hypothesis β = 0 this t-value is no longer t distributed, but ˜ t ˆ β ' ˆ β σ ˆ β . the standard normal approximation remains valid if the sample size n is large.

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

This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern