Hello, I'm working on a univariate linear regression problem with a moderate amount of data (65 points). Now, I know how to calculate the standard error for a single prediction (given that all the usual assumptions are satisfied). My question is, with this moderate amount of data, is there any way I can check how accurate is the confidence interval for a single prediction? Does a "leave-one-out" procedure offer any insight? If someone could point me to a reference I'd be very grateful. Robert
================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================