Dear all,
I am using the -ivreg- command with the "robust cluster(i) small" options. Because I have used the small option t-stasticis are specified that are robust to general forms of heteroscedasticity and within correlation. Additionally, the -test- command invokes an F-test of my specified restrictions.
Would it be better to use a Wald test of restrictions I want to test as the F-test is not valid in the presence of heteroscedasticity or is the F-test robust in the same way as the t-statistic is robust due to robust standard errors being used?
Is there a way of telling Stata to conduct a Wald test even when the default is to conduct an F-test after the 'small' option is used in the previously estimated model?
Thanks for help in advance.
Kevin
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
