bayespecon.diagnostics.bayesian_lmtests.BayesianLMTestResult¶

class bayespecon.diagnostics.bayesian_lmtests.BayesianLMTestResult(lm_samples, mean, median, credible_interval, bayes_pvalue, test_type, df=1, details=None)[source]¶

Container for Bayesian LM test results.

lm_samples[source]¶

Posterior samples of the LM statistic.

Type:¶

np.ndarray

mean[source]¶

Posterior mean of the LM statistic.

Type:¶

float

median[source]¶

Posterior median of the LM statistic.

Type:¶

float

credible_interval[source]¶

95% credible interval for the LM statistic.

Type:¶

Tuple[float, float]

bayes_pvalue[source]¶

P-value computed at the posterior mean of the LM statistic, 1 - chi2.cdf(mean, df). Under H₀ the LM statistic follows a \(\chi^2\) distribution ([Doğan et al., 2021], Proposition 1).

Type:¶

float

test_type[source]¶

Label identifying the test (e.g. "bayesian_lm_lag").

Type:¶

str

df[source]¶

Degrees of freedom for the reference \(\chi^2\) distribution. Default is 1 for single-parameter tests.

Type:¶

int

details[source]¶

Additional metadata (e.g. number of posterior draws).

Type:¶

Dict[str, Any]

__init__(lm_samples, mean, median, credible_interval, bayes_pvalue, test_type, df=1, details=None)[source]¶

Methods

__init__(lm_samples, mean, median, ...[, ...])
to_dict()	Convert the test result to a dictionary.
to_series()	Convert the test result to a pandas Series.

Attributes

details
df
lm_samples
mean
median
credible_interval
bayes_pvalue
test_type

bayes_pvalue[source]¶

credible_interval[source]¶

details = None[source]¶

df = 1[source]¶

lm_samples[source]¶

mean[source]¶

median[source]¶

test_type[source]¶

to_dict()[source]¶

Convert the test result to a dictionary.

to_series()[source]¶

Convert the test result to a pandas Series.