If you happen to only have one X variable, you could use Fit Life by X instead of Parametric Survival. (Use the Linear relationship on the Fit Life by X launch dialog.) In Fit Life by X, you have the option to **Save Probability Estimates** from the "Statistics" outline node under any of the distribution results. That option will save the estimates as well as the confidence intervals (all as formulas).

If you have more than one X variable, you'll have to stick with Parametric Survival. I don't know of a direct way to do what you're looking for in Parametric Survival, but you might be able to scriptably get enough CI values using the **Estimate Survival Probability** option in the Parametric Survival Fit red triangle menu.

Hope that helps,

Michael

Michael Crotty

Sr Statistical Writer

JMP Development