NumCosmoMathMSetTransKernGauss

A multivariate gaussian sampler.

This object subclasses NcmMSetTransKern and implements a multivariate gaussian sampler.

Implementation of a multivariate Gaussian sampler, providing a straightforward method for generating random parameter vectors with multivariate parameters. This sampler generates vectors with a normal distribution. The covariance of parameters can be configured directly using ncm_mset_trans_kern_gauss_set_cov() or by specifying individual standard deviations as parameter scales, assuming zero correlation.

Key Functionality:

• Generates random parameter vectors with multivariate parameters.
• Utilizes a multivariate Gaussian distribution for sampling.
• Allows direct setting of covariance using ncm_mset_trans_kern_gauss_set_cov().
• Supports alternative methods:
• Using ncm_mset_trans_kern_gauss_set_cov_from_scale() sets covariance using the scale property of parameters as standard deviation with zero correlation.
• Using ncm_mset_trans_kern_gauss_set_cov_from_rescale() sets covariance using the scale property of parameters times epsilon as standard deviation with zero correlation.

This implementation is particularly useful when a Gaussian sampling approach is required for generating random parameter vectors with multivariate parameters, offering flexibility in specifying covariance through direct settings or individual standard deviations.