INTRODUCTION TO ECONOMETRICS II ECO 306 NOUN 105
Example Table 1.1 Using the data in Table 1.1 above, an OLS regression of manufacturing output on GDP yields the following result (standard errors in parentheses SEE. (5700) (0.013) This implies that manufacturing accounts for $194,000 out of every $1 million increase in GDP in the cross-section. The residuals from the regression and GDP are both ranked in Table 1.2 and Di and are computed.
INTRODUCTION TO ECONOMETRICS II ECO 306 NOUN 106 Table 1.2 The sum of the latter came to 1608. The rank correlation coefficient is thus
and the test statistic is √ . This is above 2.58 and hence the null hypothesis of homoscedasticity is rejected at the 1 percent level.
3.1.3.5 The Goldfeld–Quandt Test Goldfeld and Quandt (1965) are so far attributed with the most common formal test for heteroscedasticity. The test assumes that
iu
the standard deviation of the probability distribution of the disturbance
term in observation i, is about the size of
XiIt also assumes that the disturbance term is distributed and satisfies the other Gauss–
Markov conditions. The sizeof
Xi ordersthei nobservationsinthesampleand separate regressions
are carried out for the first n'and the last
n' observations, the middle (
n–
2
n') observations being dropped completely. If heteroscedasticity is present, and if the assumption regarding
its nature is correct, the variance of
uin the last
n' observations will be more than that in the first
n' and
this will be reflected in the RSS in the two
