Method

NumCosmoMathStatsVecheidel_diag

Declaration [src]

NcmVector*
ncm_stats_vec_heidel_diag (
  NcmStatsVec* svec,
  const guint ntests,
  const gdouble pvalue,
  gint* bindex,
  guint* wp,
  guint* wp_order,
  gdouble* wp_pvalue
)

Description [src]

Applies the Heidelberger and Welch’s convergence diagnostic applying ntests Schruben tests sequentially, if ntests == 0 it will use the default 10 tests. The variable bindex will contains the smallest index where all p-values are smaller than pvalue, if pvalue is zero it used the default value of $0.05$.

If the test is not satisfied by any index bindex will contain -1 and the return vector the p-values considering the whole system.

See: - [Heidelberger (1981)][XHeidelberger1981] - [Schruben (1982)][XSchruben1982] - [Heidelberger (1983)][XHeidelberger1983].

Parameters

ntests

Type: const guint

Number of tests.

pvalue

Type: const gdouble

Required p-value.

bindex

Type: gint*

Index of the best p-values.

The argument will be set by the function.

wp

Type: guint*

Worst parameter index.

The argument will be set by the function.

wp_order

Type: guint*

Worst parameter AR fit order.

The argument will be set by the function.

wp_pvalue

Type: gdouble*

Worst parameter p-value.

The argument will be set by the function.

Return value

Type: NcmVector

A NcmVector containing the best p-values.

The caller of the method takes ownership of the returned data, and is responsible for freeing it.