ar::mean_subtracted Struct Reference

Method-specific estimation variance routines following Broersen. More...

#include <ar.hpp>

Static Public Member Functions
template<typename Result , typename Integer >
static Result	empirical_variance_zero (Integer N)
	Computes the empirical variance estimate for order zero.

Detailed Description

Method-specific estimation variance routines following Broersen.

For details see either the FiniteSampleCriteria.tex write up or Broersen, P. M. T. "Finite sample criteria for autoregressive order selection." IEEE Transactions on Signal Processing 48 (December 2000): 3550-3558. http://dx.doi.org/10.1109/78.887047.

The selection criteria routines might be sped up for floating point arguments given an appropriate digamma (psi) or Pochhammer symbol implementation. To do so with the GNU Scientific Library (GSL), e.g., try

#include <gsl/gsl_sf_psi.h>
#define AR_DIGAMMA(x) gsl_sf_psi(x)
 
#include <gsl/gsl_sf_gamma.h>
#define AR_POCHHAMMER(a,x) gsl_sf_poch(a,x)

before including this header and link the GSL with your binary. Denotes the sample mean was subtracted from a signal before estimation.

Definition at line 1460 of file ar.hpp.

---

The documentation for this struct was generated from the following file:

• ar.hpp