thanks for the reference! Calibration was discussed quite intensively here back in ~2006-2008, and Steve knows quite a lot about it.

]]>The classic regression of Y on x will have confidence intervals differing from the intervals for the inverse regression of x on Y – which you are looking for (the calibration problem).

Some methods for calculating confidence intervals for this type of problems are discussed in: Thonnard, M., (2006), Confidence Intervals in Inverse Regression. Diss. Technische Universiteit Eindhoven Department of Mathematics and Computer Science, Web. 5 Apr. 2013. .

]]>