Struct ber_interval_t¶

ClassList > ber_interval_t

A rate with its exact interval. Assert on lo , never onp_hat .More...

• #include <ber_core.h>

Public Attributes¶

Detailed Description¶

Under inverse binomial sampling the trials N are the random variable, not the errors, so the naive r/N is biased (the unbiased estimator is (r-1)/(N-1)) and the interval is the Gamma/chi-square one, [chi2_{a/2}(2r)/2N, chi2_{1-a/2}(2r)/2N]. Comparing lo against a spec is the form that cannot flake on counting noise; comparing p_hat will.

Public Attributes Documentation¶

variable conf¶

double ber_interval_t::conf;

Confidence level used.

variable errors¶

size_t ber_interval_t::errors;

r.

variable hi¶

double ber_interval_t::hi;

Upper confidence limit.

variable lo¶

double ber_interval_t::lo;

Lower confidence limit.

variable p_hat¶

double ber_interval_t::p_hat;

Unbiased point estimate (r-1)/(N-1).

variable rel¶

double ber_interval_t::rel;

Relative standard error 1/sqrt(r).

variable symbols¶

size_t ber_interval_t::symbols;

N (or bits, for a BER).

The documentation for this class was generated from the following file native/inc/ber/ber_core.h