stats

jaxns.internals.stats

Module Contents

normal_to_lognormal(mu, std)

Convert normal parameters to log-normal parameters.

Parameters:

• mu – mean of normal RV

• std – standard deviation of normal RV

Returns:

mu, sigma of log-normal RV

density_estimation(xstar, x, alpha=1.0 / 3.0, order=1)

Estimates the density of xstar given x using a trick.

Parameters:

• xstar – array of points to estimate density at

• x – array of points to estimate density from

• alpha – power law exponent

• order – order of norm to use

Returns:

density at xstar

linear_to_log_stats(log_f_mean, *, log_f2_mean=None, log_f_var=None)

Converts normal to log-normal stats. :param log_f_mean: log(E(f)) :param log_f2_mean: log(E(f**2)) :param log_f_var: log(Var(f))

Returns:

E(log(f)) Var(log(f))

effective_sample_size(log_Z_mean, log_dZ2_mean)

Computes Kish’s ESS = [sum dZ]^2 / [sum dZ^2]

Parameters:

• log_Z_mean –

• log_dZ2_mean –

Returns: