Conditional bias-corrected normal mean estimate Created 2026-10-07 Updated 2026-11-05
Consider independent stage means and , with positive known information . Continue to stage two only if . The pooled mean is , where . Write for its observed value. Its conditional mean under a candidate parameter isHere and are the standard normal probability density and cumulative distribution functions. The Conditional bias-corrected normal mean estimate solves . This inverts the selection-shifted mean instead of ignoring the fact that the first-stage mean exceeded a threshold.
For the upper-tail inverse Mills ratio , . The variance of a truncated standard normal variable gives , hence . Since , is strictly increasing and ranges over the whole real line; the root exists uniquely. Differentiating the joint normal likelihood conditional on continuation shows that the same root maximizes that conditional likelihood. Inverting a conditional mean does not itself make an estimator conditionally unbiased. The separate Rao-Blackwell estimator after interim selection provides that property in the equal-information setting.
Past exam of the mathematics course of the University of Cambridge 2013 iii Paper 32 3 b iii Solution Created 2026-10-03 Updated 2026-10-07
Both alternatives can be made explicit. For a conditional bias-corrected normal mean estimate, let be the observed pooled estimate and , with . Invert its conditional mean:This can be solved by bracketing or by Newton iteration . Since ,The variance of a standard normal conditional on exceeding is , strictly between zero and one. Positivity follows from nondegeneracy; the upper bound follows because the conditional mean exceeds the truncation threshold . Hence , so the equation has at most one root. As , ; as , the truncated stage-1 mean approaches its threshold and , so a root exists for every finite . Equivalently, differentiating the conditional likelihood divides the ordinary likelihood by and gives the same score equation. This conditional-likelihood correction is not exactly conditionally unbiased merely because it inverts a mean.
For an illustration, take , , and . The equation is . Since and , the corrected estimate lies between zero and ; numerical solution gives , below the selected ordinary estimate.
For the uniform minimum variance conditionally unbiased estimator, put , , and . The fresh estimate is conditionally unbiased because it is independent of continuation. LetBefore selection, conditional Gaussian calculations give . Conditional on as well, this normal variable is truncated below , soUsing , Rao-Blackwellization therefore givesIts conditional expectation is , and its conditional variance cannot exceed that of . To justify uniform minimum variance, the joint conditional density of is a base density on multiplied by . Thus is a complete sufficient statistic in the one-parameter conditional exponential family, whose natural parameter ranges over an open real interval. The Lehmann–Scheffé theorem proves the claim. The orthogonal pooled arm-average statistic is independent of the entire difference process and carries the nuisance common mean. Together with , it gives a complete sufficient statistic in the selected two-parameter normal family, with an open natural-parameter space. Thus allowing that nuisance statistic does not improve the conditional unbiased estimate of the difference. For the same illustration, , and give . It differs from the conditional-likelihood estimate because exact conditional unbiasedness is a different criterion.