heteroscedastic bivariate distribution with linear regression - prediction interval

Dear forum,

which is the most suitable method to get the prediction interval of a
bivariate normal distribution which is consistent with a linear model y = ax
+ b?

I assume it is gls + predict. Am I correct? I'm rather new to R. 
Is there some reliable sample code for that problem?

Thank you
best regards

--
View this message in context: http://r.789695.n4.nabble.com/heteroscedastic-bivariate-distribution-with-linear-regression-prediction-interval-tp3981793p3981793.html
Sent from the R help mailing list archive at Nabble.com.