Consistent estimator

Consistency is a concept of mathematical statistics. He presses a minimum requirement of estimators and statistical tests so that they apply as practicable ( additional requirements: ( asymptotic ) unbiasedness, ( asymptotic ) efficiency and sufficiency). Spoken heuristic estimators are consistent if an increase in the sample size means that the estimator is closer to the true value of the parameter to be estimated. Ideally, an infinite number of observations estimated and true parameters should be identical.

Formal consistency corresponds to an estimation function for (or for a functional of ), generally of the stochastic convergence. Analog hot tests consistent if the probability of a Type II error stochastically converges to 0. It is sometimes also between weak consistency (equivalent to consistency in the above sense ), strong consistency (equivalent to almost sure convergence ) and consistency in the root mean square (equivalent to convergence in the quadratic mean ) differed. All mentioned here convergence types are defined and classified in the survey article for convergence in the stochastic.