In statistical hypothesis testing, the error exponent of a hypothesis testing procedure is the rate at which the probabilities of Type I and Type II decay exponentially with the size of the sample used in the test.
For example, if the probability of error
of a test decays as
is the sample size, the error exponent is
Formally, the error exponent of a test is defined as the limiting value of the ratio of the negative logarithm of the error probability to the sample size for large sample sizes:
Error exponents for different hypothesis tests are computed using Sanov's theorem and other results from large deviations theory.
Consider a binary hypothesis testing problem in which observations are modeled as independent and identically distributed random variables under each hypothesis.
denote the observations.
denote the probability density function of each observation
under the null hypothesis
denote the probability density function of each observation
under the alternate hypothesis
In this case there are two possible error events.
Error of type 1, also called false positive, occurs when the null hypothesis is true and it is wrongly rejected.
Error of type 2, also called false negative, occurs when the alternate hypothesis is true and null hypothesis is not rejected.
The probability of type 1 error is denoted
and the probability of type 2 error is denoted
In the Neyman–Pearson[1] version of binary hypothesis testing, one is interested in minimizing the probability of type 2 error
subject to the constraint that the probability of type 1 error
is less than or equal to a pre-specified level
In this setting, the optimal testing procedure is a likelihood-ratio test.
[2] Furthermore, the optimal test guarantees that the type 2 error probability decays exponentially in the sample size
[3] The error exponent
is the Kullback–Leibler divergence between the probability distributions of the observations under the two hypotheses.
This exponent is also referred to as the Chernoff–Stein lemma exponent.
In the Bayesian version of binary hypothesis testing one is interested in minimizing the average error probability under both hypothesis, assuming a prior probability of occurrence on each hypothesis.
denote the prior probability of hypothesis
In this case the average error probability is given by
In this setting again a likelihood ratio test is optimal and the optimal error decays as
represents the Chernoff-information between the two distributions defined as