Confidence intervals based on one or more observations
Nelson M. Blachman, Robert E. Machol · IEEE Transactions on Information Theory · 1987
On the basis ofn\geq 2observations, confidence limits of the formX̄ \pm tS/ \sqrt{n}are constructed for the location (e.g., the median) of any distribution of known form with unknown location and dispersion (scale), whereX̄andSare the sample mean and "unbiased" standard deviation. Particular attention is given to the values oftneeded for the Cauchy and uniform distributions. The lattertsuffices for any (unknown) symmetric unimodal distribution ift \geq n - 1. A table compares these values oftforn=2,3,4, and5with those for the normal case, which are derived here very simply and are identical with those found by "Student." We are also able to include the case of a single observation(n=1), where confidence intervals of various forms are made just wide enough for the least favorable dispersion. They, therefore, include the true location with at least but, in general, not exactly the desired probability; these intervals involve a predetermined value that plays a role reminiscent of but quite different from that of the prior distribution that would enter into a Bayesian analysis. In addition, upper confidence limits for the dispersion are constructed forn \geq 1.