Suppose, that, in some regression model, you want pointwise bootstrap
bca (bias-corrected, accelerated) confidence intervals for the predicted values.
In practice, the regression model might specify that the expected value
of the outcome variable is some complicated function of the predictors.
But to make this concrete, consider the following simple situation:
use http://www.ats.ucla.edu/stat/stata/examples/chp/P004.dta, replace
keep fat protein
order fat protein
regress protein fat
predict PRED
To get bias-corrected, accelerated bootstrap confidence intervals merely for
the regression coefficients, one does the following:
bootstrap, bca reps(2000): regress protein fat
estat bootstrap, bca
But is there a simple way to obtain a bootstrap confidence interval for
the predicted values? For instance, can one do it with a one-line call
to the "bootstrap" command? Or if it's not simple, how does one do it?
Thanks for any info
Jake Wegelin
*
* 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/
