P. Rilstone et M. Veall, USING BOOTSTRAPPED CONFIDENCE-INTERVALS FOR IMPROVED INFERENCES WITH SEEMINGLY UNRELATED REGRESSION EQUATIONS, Econometric theory, 12(3), 1996, pp. 569-580
The usual standard errors for the regression coefficients in a seeming
ly unrelated regression model have a substantial downward bias, Bootst
rapping the standard errors does not seem to improve inferences, In th
is paper, Monte Carlo evidence is reported which indicates that bootst
rapping can result in substantially better inferences when applied to
t-ratios rather than to standard errors.