Confidence Interval

A confidence interval is a range of values, calculated from sample data, constructed by a procedure guaranteed to contain the true value of an unknown population parameter a stated proportion of the time were the sampling repeated indefinitely, commonly ninety five percent. Jerzy Neyman introduced the concept in his 1937 paper Outline of a Theory of Statistical Estimation Based on the Classical Theory of Probability, giving frequentist statistics a way to express the uncertainty of an estimate without adopting the prior probabilities Bayesian methods require. It is among the most widely reported statistics in the sciences alongside the p-value, though its precise meaning, a statement about the reliability of the estimating procedure rather than a probability about the parameter itself, is also among the most widely misunderstood.